Done, your profile is created.Finish your profile by filling in the following fields
Forgot Password Earn Money,Free Notes
Password sent to your Email Id, Please Check your Mail
Updating Cart........ Please Wait........
e‐Commerce Notes 24
Part Two of Two
E‐Business Introduction .................................................................................................................................................... 2
E‐Business Strategies ...................................................................................................................................................... 15
Integration of Applications ............................................................................................................................................. 24
E‐Commerce Infrastructure ............................................................................................................................................ 46
Vidyalankar Institute of Technology Page 1
EBusiness vs. Ecommerce
While some use e‐commerce and e‐business interchangeably, they are distinct concepts. Electronic business,
commonly referred to as "e‐Business" or "e‐business", may be defined as the application of information and
communication technologies (ICT) in support of all the activities of business. Commerce constitutes the exchange of
products and services between businesses, groups and individuals and can be seen as one of the essential activities
of any business. Electronic commerce focuses on the use of ICT to enable the external activities and relationships of
the business with individuals, groups and other businesses.
E‐Commerce Is a particular form of e‐Business. Electronic business methods enable companies to link their internal
and external data processing systems more efficiently and flexibly, to work more closely with suppliers and partners,
and to better satisfy the needs and expectations of their customers. Compared to e‐Commerce, e‐Business is a more
generic term because it refers not only to information exchanges related to buying and selling but also servicing
customers and collaborating with business partners, distributors and suppliers.
E‐Business encompasses sophisticated business‐to‐business interactions and collaboration activities at a level of
enterprise applications and business processes, enabling business partners to share in‐depth business intelligence,
which leads, in turn, to the management and optimization of inter‐enterprise processes such as supply chain
management. More specifically, e‐Business enables companies to link their internal and external processes more
efficiently and flexibly, work more closely with suppliers and better satisfy the needs and expectations of their
In practice, e‐business is more than just e‐commerce. While e‐business refers to more strategic focus with an
emphasis on the functions that occur when using electronic capabilities, e‐commerce is a subset of an overall e‐
business strategy. E‐commerce seeks to add revenue streams using the World Wide Web or the Internet to build and
enhance relationships with clients and partners and to improve efficiency using the Empty Vessel strategy. Often, e‐
commerce involves the application of knowledge management systems.
E‐business involves business processes spanning the entire value chain: electronic purchasing and supply chain
management, processing orders electronically, handling customer service, and cooperating with business partners.
Special technical standards for e‐business facilitate the exchange of data between companies. E‐business software
solutions allow the integration of intra and inter firm business processes. E‐business can be conducted using the
Web, the Internet, intranets, extranets, or some combination of these.
Basically, electronic commerce (EC) is the process of buying, transferring, or exchanging products, services, and/or
information via computer networks, including the internet. EC can also be benefited from many perspective
including business process, service, learning, collaborative, community. EC is often confused with e‐business.
In e‐commerce, information and communications technology (ICT) is used in inter‐business or inter‐organizational
transactions (transactions between and among firms/organizations) and in business‐to‐consumer transactions
(transactions between firms/organizations and individuals).
In e‐business, on the other hand, ICT is used to enhance one’s business. It includes any process that a business
organization (either a for‐profit, governmental or non‐profit entity) conducts over a computer‐mediated network.
A more comprehensive definition of e‐business is: “The transformation of an organization’s processes to deliver
additional customer value through the application of technologies, philosophies and computing paradigm of the new
Vidyalankar Institute of Technology Page 2
Three primary processes are enhanced in e‐business:
• Production processes, which include procurement, ordering and replenishment of stocks; processing of
payments; electronic links with suppliers; and production control processes, among others;
• Customer‐focused processes, which include promotional and marketing efforts, selling over the Internet,
processing of customers’ purchase orders and payments, and customer support, among others
• Internal management processes, which include employee services, training, internal information‐sharing,
video‐conferencing, and recruiting. Electronic applications enhance information flow between production
and sales forces to improve sales force productivity. Workgroup communications and electronic publishing
of internal business information are likewise made more efficient.
E‐Business goes far beyond e‐commerce or buying and selling over the Internet, and deep into the processes and
cultures of an enterprise. It is the powerful business environment that is created when you connect critical business
systems directly to customers, employees, vendors, and business partners, using Intranets, Extranets, ecommerce
technologies, collaborative applications, and the Web.
E‐business is a more strategic focus with an emphasis on the functions that occur when using electronic capabilities
while E‐commerce is a subset of an overall e‐business strategy. E‐commerce seeks to add revenue streams using the
World Wide Web or the Internet to build and enhance relationships with clients and partners and to improve
efficiency while Electronic business methods enable companies to link their internal and external data processing
systems more efficiently and flexibly, to work more closely with suppliers and partners, and to better satisfy the
needs and expectations of their customers.
E‐Business is at the enterprise application level and encompasses sophisticated b2b interaction and collaboration
activities. Enterprise Application Systems such as ERP, CRM, SCM form an integral part of e‐Business strategy and
Critical Factors with respect of eBusiness
E‐Business supports business processes along the entire value chain: Electronic purchasing (E‐Procurement), SCM
(Supply Chain Management), Processing orders electronically, Customer Service Co‐operation with business
One of the objectives of e‐Business is to provide seamless connectivity and integration between business processes
and applications external to an enterprise and the enterprise’s back office applications sucha as billing, orger
processing, accounting, inventory and receivables, and services focused to total supply chain management and
partnership including product development, fulfillment, and distribution. In this respect, e‐Business is much more
To succeed in e‐Business it is crucial to combine technological developments with corporate strategy that redifines a
company’s role in the digital economy while taking into account its various stakeholders. It is imperative to
understand the issues, evaluate the options, and develop technology orientation plans. An e‐Business strategy helps
organizations identify their e‐Business concerns, assess their information needs, analyze to what degree existing
systems serve these objectives, pinpoint specific improvements, determine the development stages of e‐Business
solutions and attain concrete and measurable results. Thus, it is clear that e‐Business solutions are not only about
A classic example is SAP systems integrations for any organization. This itself is taken up as a project and executed
with great attention to detail. A minute logical error in interpretation of the firm’s objectives could result in the
entire system being re‐worked from scratch.
Vidyalankar Institute of Technology Page 3
E‐Business allows for redefinition of value, competitiveness and the very nature of transactions and it affects all
areas of an organization. It is crucial to combine technology and business strategy while taking into account various
An E‐business Strategy helps to
• Identify e‐business concerns
• Assess info needs
• Analyze existing systems
• Improvements required in existing systems
• Determine the stages of development of solutions
• Attain concrete and measurable results.
Characteristics of eBusiness
To emphasize, e‐Business is not simply buying and selling but encompasses the exchange of many kinds of
information, include online commercial transactions. E‐Business is about integrating external company processes
with an organization’s internal business processes; as such, a variety of core business processes could exploit an e‐
These include among others: ‐
• Collaborative Product Development
• Collaborative Planning, Forecasting and Replenishment
• Procurement and Order management
• Operations and Logistics
Collaborative Product Development
This is one of the fastest growing technologies in engineering with some form of solutions being implemented in a
range of industries such as automotive, aerospace, agricultural machinery etc. It contributes towards making
products in a short time span while maintaining quality and reducing cost.
It also aids in maximizing time‐to‐market benefits while maintaining control over product development information.
By integrating design and testing cycles of products with those of suppliers, a firm can shorten the complete cycle of
its products. This clearly, reduces the total cost of the product cycle, even more importantly, it reduces the time
that is needed to bring products to the marketplace. Collaborative product development solutions offer ERP
integration and SCM.
Collaborative Planning, Forecasting and Replenishment
This is a process in which Manufacturers, Distributors and Retailers work together to plan, forecast and replenish
products. In e‐Business relationships collaboration takes the form of sharing information that impacts inventory
levels and merchandise flow.
Collaboration points: sales forecasts, inventory requirements, manufacturing and logistic lead times, seasonal set
schedules, new/remodel storage plans, promotional plans etc
Goal: To get the partners to work together to improve lower supply cycle times, improve customer service, lower
inventory costs, improve inventory levels and achieve better control of planning activities
Procurement and Order management
Electronic procurement or E‐Procurement can achieve significant savings and other benefits that impact the
customer. To support procurement and order management processes, companies use an integrated electronic
ordering process and other online resources to increase efficiency in purchasing operations.
Vidyalankar Institute of Technology Page 4
Benefits: cost savings, better customer service by controlling the supply base, negotiating effective buying
preferences, and streamlining the overall procurement process.
Logistics is that part of the supply chain process that plans, implements and controls the efficient, effective flow and
storage of goods, services and related information from the point of origin to point of consumption in order to meet
customer requirements. To make this happen, transportation, distribution, warehousing, purchasing order
management functions must work together. Logistics in the e‐Business era is all about Collaboration ‐ the sharing of
critical and timely data on the movement of goods as they flow from raw material, all the way to the end‐user.
Operations and Logistics processes are based on open communication between networks of trading partners where
integrated processes and technology are essential for high performance logistics operations. These solutions help
manage the logistics process between buyers and suppliers, while eliminating costly discrepancies between purchase
order, sales order and shipping information. By eradications these variances and inconsistencies improvements in
the supply chain may result from the elimination of mixed shipments and shipment discrepancies, and the reduction
of inventory carrying costs for the customer. At the same time this increases customer satisfaction through
improved delivery reliability and improved efficiencies in receiving operations.
Furthermore, there are critical elements to e‐business models as well. They are as follows:
• A shared digital business infrastructure, including digital production and distribution technologies
(broadband/wireless networks, content creation technologies and information management systems), which
will allow business participants to create and utilize network economies of scale and scope.
• A sophisticated model for operations, including integrated value chains‐both supply chains and buy chains.
• An e‐business management model, consisting of business teams and/or partnerships;
• Policy, regulatory and social systems ‐ i.e., business policies consistent with e‐commerce laws, tele‐
working/virtual work, distances learning, incentive schemes, among others.
• Ease of Automated Processing ‐ A payer can now cheaply and easily automate the generation and
processing of multiple payments with minimal effort. Previously, the dependency upon banks to handle
most payments and the lack of a cheap, ubiquitous communications technology made automation of
payment processes expensive and difficult to establish.
• Immediacy of result ‐ Payment immediacy occurs because automation and the ability for the intermediate
systems and providers to process payments in real‐time. With the more manual, paper‐based systems there
was always a time delay due to the requirement for human intervention in the process.
• Openness and accessibility ‐ The availability of cheap computing and communications technology and the
appropriate software enables small enterprises and individuals to access or provide a range of payment
services that were previously only available to large organizations via dedicated networks or the
transactional processing units of banks.
• Loss of collateral information ‐ The new technology dispenses with, or alters, collateral information
accompanying transactions. This information has traditionally been part of the transaction, and has been
relied upon by the transacting parties to validate individual payments.
• Collateral information can be defined as information:
• Which is not essential to the meaning and intent of a transaction;
• Which is typically incidental to the nature of the communications channel over which the transaction is
conducted; but nevertheless provides useful contextual information for one or more of the parties to the
• Collateral information can include many things ranging from tone of voice in a telephone call to the business
cards and letterheads and apparent authority of the person with whom you are dealing.
Vidyalankar Institute of Technology Page 5
• Globalizzation ‐ Glob balization, orr the minimization of geographical factors in maaking payments, has bee en an
obvious aspect of the new paayments sysstems. Its afffect is upon areas succh as size o of the paym ments
marketp place, uncerrtainty as to o legal jurissdiction in tthe event o of disputes, location an nd availabilitty of
transacttion trails, an nd the abilitty of a paym ment scheme e to rapidly aadapt to reggulatory regimes impose ed by
one country by moving to another.
• New business mode els ‐ New bussiness models are being developed to exploit the e new payme ent technolo ogies,
in particcular to addrress or take advantage o of the disinttermediation n of custome ers from trad ditional paym ment
providerrs such as baanks. In this context, dissintermediation is where e the techno ology enabless a third parrty to
interven ne between the customer and the banking systtem, effectivvely transferrring the customer’s tru usted
relationsship with the e bank to the e new party.
Eleements of an eBusiiness solu ution
The vision of e‐ ‐Business is that enterp prises will haave access tto broad ran nge of tradiing partnerss to interactt and
collaaborate with h and not o only buy and d sell more efficiently. Also, it is e expected thaat e‐Business will contriibute
towards the agillity of busine ess organizattions and witth that to reaching highe er levels of ccustomization. In this way, an
orgaanization can n maximize ssupply chain efficiency, im mprove customer service e and increase profit margins. Hence e, the
need d to make m mission criticaal processes:
Inve entory, Acco ounting, Man nufacturing and Custom mer Support: These, musst be able to o interact witth each othe er by
beco oming web‐e enabled. This is achieved d by ERP, CR RM and othe er systems b by making usse of distribu uted applicattions
thatt extract dataa and launch h business processes acro oss many or aall of the abo ove processe es.
The key elementts of an e‐Bu usiness solutiion are:
1. Custome er Resource managemen nt(CRM)
2. Enterpriise resource planning (ER RP)
3. Supply C Chain Managgement (SCM M)
4. Knowled dge Management
5. e‐Marke ets
Vidy yalankar Ins stitute of Technology Pa age 6
Customer relationship management (CRM )
CRM systems are “front‐office” systems which help the enterprise deal directly with its customers. CRM (definition)
is the process of creating relationships with customers through reliable service automated processes, personal
information gathering, processing and self‐service through the enterprise in order to create value for customers.
There are 3 categories of user applications under CRMs:
• Customer‐facing applications:
Applications which enable
customers to order products and
• Sales‐force facing applications:
Applications that automate some of
the sales and sales‐force
management functions, and support
dispatch and logistic functions.
• Management‐facing applications:
Applications which gather data from
previous apps and provide
management reports and compute
Return on relationships(RoR) as per
company’s business model
Enterprise Resource Planning (ERP)
ERPs are often called “back‐office” systems. ERP systems are management information systems that integrate and
automate many of the business practices associated with operations or production aspects of a company. ERP
software can aid in control of many business activities such as sales, delivery, production, billing, production,
inventory, shipping, invoicing and accounting.
A typical ERP system is designed around these 4
primary business procedures: ‐
• Production: manufacturing, resource planning
and execution process
• Buying a product: procurement process
• Sales of a product and services: customer
order management process
• Costing, paying bills, and collecting:
financial/management accounting and
Supply Chain Management (SCM)
Supply chain (definition) is a network of facilities and distribution options that perform the functions of procurement
of materials, transformation of these materials into intermediate and finished products, and distribution of these
finished products to customers. SCM deals with the planning and execution issues involved in managing a supply
Vidyalankar Institute of Technology Page 7
Supply chain has 3 main parts
• Supply side: concentrates on
how, where from, and when raw
materials are procured and
supplied to manufacturing.
• Manufacturing side: converts raw
materials to finished products.
• Distribution side: ensures that
finished products reach the final
customers through a network of
distributors, warehouses and
This relates to the identification and analysis of available and required knowledge assets and related processes.
Knowledge assets encompass two things Information and Experience. Knowledge assets comprise of all knowledge
that a business has or needs to have in order to generate profits and add value.
Knowledge management includes the subsequent planning and control of actions to develop both the knowledge
assets and the processes to fulfill organizational objectives. Knowledge is a strong denominator of a business model
and determines business competencies especially when unique to the business and so must be kept in‐house.
E‐Market is an electronic meeting place for multiple buyers and sellers providing many participants with a unified
view of sets of goods and services, enabling them to transact using many different mechanisms. An e‐Market uses
Internet technology to connect multiple buyers and suppliers.
EBusiness Roles and their challenges
• There are two main roles in the E‐business scenario:
o The Buyer: Buyers are organizations that purchase goods and services directly from Suppliers.
o The Supplier: Suppliers are organizations that market and sell goods or services directly to buyers or
indirectly through diverse sales channels including Web‐based procurement systems and electronic
• Suppliers typically provide buyers with web‐based services necessary for completing e‐Business transactions.
• Buyers (customers) can thus review product information, receive customer service, ordering services and
customization support facilities an can submit or modify orders.
• An additional role is that of Market Makers that are third party organizations that run e‐markets.
• Each role has distinct business and technical challenges, but they all coalesce around a common point.
• For buyers as well as for suppliers, the primary challenge is the ability to reach a critical mass of trading
partners and transaction volume to sustain their business.
• For suppliers especially, the following challenges exist:
o Managing multiple selling channels, based on various technologies, protocols, data formats, and
standard business processes.
o Having the ability to take multiple types of orders once the customer has decided to conduct e‐
Business –enabled order management through the various selling channels.
o Having the ability to differentiate and customize products and services from other suppliers, and
offering them through the various selling channels.
o Having the ability to adapt and grow the e‐Business without incurring drastic technology changes,
Vidyalankar Institute of Technology Page 8
o And sweeping changes in the business process, or radical new investments.
• To meet the needs of buyers and suppliers, e‐Business strategy and solutions must be built on the following
o Empowering suppliers buyers:
o Enabling suppliers of all sizes:
• Identify/measure quantifiable business objectives: companies must accurately measure the impact an e‐
Business initiative has on their business processes and decide whether this initiative is worth pursuing and
has sustainable long‐term effects
• Ensure organizational/operational flexibility: Enterprises must reposition themselves in their mission,
structure and execution to prosper in a substantially more dynamic environment.
• Rethink entire company supply chains: companies must rethink their entire supply chains in order to
optimize performance and value as they seek to better integrate with suppliers and customers, share
information, inter‐link processes, and outsource manufacturing logistics systems and maintenance activities.
• Transform the company to a process‐centric form: Companies must be conceptualized as a set of business
processes with more emphasis on maximizing the efficiency of processes rather than departmental or
• Define Business processes: companies must create models of existing processes and interactions
determining the relevant events, time frames, resources and costs associated with business processes,
hence making them well‐defined and measurable
• Understand Security requirements: the breadth of access and interaction requirements of a e‐Business
solution requires the ability to provide controlled and focused access by all the users.
• Align business organizations with a flexible IT architecture: in response to demands for end to end e‐
Business solutions, companies are expanding their applications to include enhanced integration capabilities.
This includes integration of business processes at varied levels from applications and data across(and within)
• Establish ubiquity within standards: None of the many integration technologies available from various IT
vendors has achieved complete coverage. These do work within organizations but not across global
enterprises and between separate enterprises. Attempts are made to establish open standards for
• A number of business and tech. driven requirements are compelling forces that enable successful
development deployment of integrated end‐to‐end e‐Business applications. Some of these are:
o Efficient business process management technology
o Efficient b2b communication
o Efficient enterprise application integration technology
Other categorizations view the problem differently.
A more basic approach to viewing e‐Business requirements is as follows: ‐
• Trust ‐ The biggest requirement for running a successful e‐business is trust. In this age of Facebook and
MySpace, online merchants may think that privacy of a customer's information isn't important, but the
opposite is true.
o Thus, businesses must be trustworthy to operate online. Consumers will not simply give their
financial information to just anyone, so a site will lose business if consumers do not feel comfortable
that it is a reliable, upstanding company.
Vidyalankar Institute of Technology Page 9
o Companies must have comprehensive privacy policies and stick with them. Another good idea is to
get digital certificates and TRUSTe seals, which are awarded by third‐party organizations after they
research the legitimacy of an online website.
o Such awards put consumers' minds at ease. Finally, even if an e‐business does all this, it must also be
trustworthy in the sense of fulfilling its promises: be up front with consumers about pricing and
think about when dealing with the Internet and e‐business.
You should fully understand how your website fits into privacy law requirements.
If your website collects personal information, you should develop a proper and legally
still need to develop and post a policy.
Online profiling may require the consent of the individual depending on the circumstances.
Keep in mind that people do look for privacy policies so, without a policy, you may lose
can serve a marketing function as well, allowing you to attract and retain customers who
otherwise might not be as inclined to deal with you.
Do not create a policy and then fail to follow it precisely. This is an invitation for disaster,
including not only possible legal problems, but also injury to your reputation and goodwill.
o It is important to not just let the policy sit once it has been posted. It should be revisited regularly to
determine whether or not it is still accurate and to evaluate whether or not it should be revised to
assist you in your business goals and objectives.
• Strategy ‐ E‐commerce merchants must also have a strategy to succeed in the online marketplace. Many
people start websites because they think it is a quick and easy way to make cash, but in fact it takes a much
greater investment than most people expect.
• Therefore, before launching a site, businesses must have strategies to handle issues large and small:
o How consumers will place orders,
o How deliveries will be made,
o How customer service issues will be handled
o More broadly, how much do owners expect to earn over a certain period, how will consumers find
the site, and how will success be judged.
o Online merchants without strategies will soon be overwhelmed by such issues.
• Suitability ‐ Finally, merchants must decide if their products are suitable for the web. Requirements for
successful e‐businesses concern the goods and services themselves:
o Can they be delivered quickly and cheaply
o Do they appeal to people outside a small geographic area
o Will going online save money
o Will the benefits outweigh the costs
Vidyalankar Institute of Technology Page 10
• Technological Requirements:
o Achieving Real ‐ Time Flexibility: In theory, digital things are easier to change than physical things. It
is faster to edit a memo using a word processor than a typewriter (and you don’t get ink on your
o But when programming is required to change content or access policies, maintaining a complex Web
site can range from onerous to impossible. Market factors change in real time, and so must the logic
and content of an e‐Business site.
o To achieve this vision, the next generation of e‐Business systems must provide a framework for
automated information exchange between all the stakeholders in a business.
o These new frameworks are designed for flexibility so companies can change content and business
logic in real‐time to meet changing business needs and market conditions.
o This adaptability comes from a set of core services, common to all applications, which enable rapid
deployment of new applications and new information and which work together to create a
compelling, unified e‐Business environment.
o An e‐Business framework must include packaged, ready‐to‐deploy services for:
o An Architecture For e‐Business ‐ As e‐Business moves beyond simple transactions to encompass all
the complex processes through which a company provides value, information systems must
orchestrate the function of enterprise applications and information resources for total information
flow. And they must empower business people with the tools to manage content publishing,
delivery, and access, so that business results don’t depend on the IT department’s programming
o Three ‐ Tier Object ‐ Centered Design ‐ To achieve true, real‐time e‐Commerce, next‐generation e‐
Business systems must be built around a 3‐tier application paradigm with a clear abstraction and
true separation of user interface presentation, business logic, and content. Separation and
abstraction of these layers is achieved through the use of business objects, particularly in the middle
o When separating presentation, application, and Data Logic three things must be considered:
• User Interface ‐ The user interface must support a variety of interface mechanisms,
including Web browsers for users, business managers, designers and desktop applications
• Business Logic ‐ The middle tier must not only implement and execute business logic, it must
also provide the framework of services that enable e‐Business, including security services,
Vidyalankar Institute of Technology Page 11
transaction services, and caching, pooling, and other load balancing services to improve
overall system performance.
• Content ‐ The content layer includes corporate databases, document stores and other
o All Objects Are Not Created Equal ‐ The overall architecture of an e‐Business system is important, but
proper abstractions achieved through object technology are the foundation of a flexible e‐Business
system. Correct separation of presentation, business logic, security functions, and content determines
the flexibility of the system and the pace and effectiveness of e‐Business processes. To deliver truly
dynamic, real‐time communication, these relationships must be established on a per‐transaction basis,
as each page is assembled for delivery to a user e‐Business processes lend themselves to this kind of
o Bringing Order To Content Management ‐ As companies move more of their business processes onto
the Web in search of greater sales or efficiency, Web sites are growing in size and complexity. Static Web
sites often consist of hundreds or even thousands of Web pages, and tens of thousands of lines of code.
Multi‐media sites are becoming the standard, with everything from sophisticated graphics and
animation to audio and video. Enterprise Web sites must integrate multiple applications from the back‐
office to the supply and sales chain, while maintaining security and the integrity of business information.
As sites become larger and more complex, traditional Web publishing systems, with their hard‐coded
Web page content, become unmanageable. Content creators swamp programmers with requests for
new Web pages, the approval process bogs down, and users no longer have access to current content.
Vidyalankar Institute of Technology Page 12
o Dynamic Web Environment ‐ The graphical layouts used in next‐generation e‐Business systems are more
“intelligent” and manageable than the templates used in traditional Web publishing. While both control
placement of graphic elements, style, etc., templates access content through business logic hard‐coded
into the body of the page. Pages with different content, however similar, require different source files.
Layouts, on the other hand, are a next‐generation approach that does not embed business logic in the
presentation objects. A layout controls only style and placement of elements on the page. The logic that
determines content is separate from the layout, and can be changed and maintained independently.
o Content Management Tools ‐ Content Management tools must enable content to grow and change at
o Team Content Development ‐ Publishing content to the Web and extending the functionality of a site
takes a whole team: developers to build site structure and implement business rules, designers to create
page layouts and define a consistent look and feel, and business managers to define business rules and
o Collaboration Across The Extended Enterprise ‐ Publishing and managing Web content typically involves
an approval process and some administrative work.
o Centralized Rules ‐ Based Content Management: Anyone should be able to manage site content simply
by defining a few document characteristics when a document is published. With characteristics such as a
document’s type (for example, “data sheet”), format, and activation/expiration dates in place, links to
the document can be automatically populated throughout the site, and document visibility and
“document migration” can be automated.
o Customized Content Delivery
o Pervasive Personalization
o Knowledge‐Based Personalization ‐ Effective personalization depends on the ability to customize a
user’s experience based on a rich, centrally stored user profile: in essence, a knowledge base that
consists of user information and expertise on how to apply that information. This kind of knowledge
base cannot be bolted onto a brochure ware Web site. The ability to gather and apply user‐related
knowledge must be integrated into the e‐Business system from day one, so that information can be
contributed, shared, and leveraged by all the applications in the system.
o User profiles are the crown jewels of an e‐Business strategy. The quality of profiles determines the
degree to which the user experience can be personalized. Profiles can and should be built through both
explicit and implicit mechanisms.
o Inclusive Security
o Scalability To Compete ‐ As more and more processes are adapted to e‐Business, a Web site may grow
to support thousands of users, millions of documents and millions of transactions each day. An e‐
Business system must have the power to perform fast and reliably, as a business grows, while delivering
the dynamic, personalized content necessary to achieve business goals.
o Enterprise Integration and Transaction Monitors ‐ Any business process can be “Web‐ified” with a CGI
interface or a few server pages. But isolated, e‐Business is about providing new value by doing business
in a fundamentally new way. Integration is the goal and the heart of e‐Business: integrating and exposing
applications and content in a personalized way to speed, scale and improve business processes and to
engage, involve, and build lasting relationships with customers and business partners.
o Transaction management guarantees that users have a consistent view of business information. For
example, the e‐Business system should prevent the customer from completing an order based on one
price, then being charged based on the new price. Transaction management also prevents inaccurate
results based on system failures (e.g., the system goes down and loses an order but continues to process
the billing using already‐transmitted credit card information). Robust system logs can help coordinate
updates across multiple data sources from multiple vendors or roll back changes in case of system
Vidyalankar Institute of Technology Page 13
o Delegated System Management ‐ e‐Business systems are distributed by their very nature, coordinating
information sharing among applications, business functions and departments, and partners up and down
the supply chain. Bringing business processes to the Web increases the complexity of the e‐Business site,
and growing and changing numbers of users and applications increase the complexity of managing a site.
No centralized IT department could effectively maintain current accounts or access privileges for all
users, inside and outside the company. Most Web sites today are not sophisticated enough to reach this
roadblock, but as businesses open and extend their processes via the Web, system manageability will
become an increasingly serious issue.
o Time to Market –Time to market must be minimal as delays may result in losing the benefit of e‐
Impacts of eBusiness
• Improved operational efficiency and productivity: by eliminating operational waste and automation of
inefficient business practices, organizations can realize productivity gains
• Reduction in operating costs and costs of goods and services: by connecting directly with suppliers and
distributors, organizations can realize more efficient processes that result in reduced units of cost for
products or services and lower prices to customers while achieving economies of scale.
• Improved competitive position: global reach, rapid growth, efficient reduction of product time to market
and optimization of product distribution channels all contribute to superior competitive position.
• Penetration into new markets through new channels: with e‐Business location is of no consequence when it
comes to reaching customers.
• Improved communication, information and knowledge sharing: alignment of key supply chain partners with
an organization’s internal strategies helps exploit their expertise and knowledge, hence creating opportunity
to secure long‐term business by embedding their process and procedures in those of their customers’ supply
• Harmonization and standardization of process
• Improved internal information access
• Improved relationships with suppliers and improved customer service
Inhibitors of e‐Business
• Management/Strategy issues
o e‐business strategy
o Organizational changes required by e‐business
o Management attitudes and organizational inflexibility
• Cost/financing issues
o Costs of implementation
o Calculating the Return on Investment (ROI)
• Security and Trust Issues
• Legal Issues
o Few companies are familiar with the rules and regulations that apply to an online environment.
This leads to Uncertainty.
o Different strokes for different folks
• Technological Concerns
o Integration Issues
• Arguments against Investment
o Uncertainty Fear
Vidyalankar Institute of Technology Page 14
What is an EBusiness Strategy
• E‐Business has triggered new business models, strategies and tactics that are made possible by the internet
and other related technologies.
• In order to compete in the marketplace, it is essential for organizations to establish strategies for the
development of an e‐business.
• E‐Business strategy can be viewed via two different viewpoints, which are explained below.
• One view defines strategy as plans and objectives adopted to achieve higher‐level goals.
• In that sense, a strategy is developed to achieve a goal like implementing organizational change, or a large
software package such as an ERP‐system.
• Strategy may also relate to plans concerning the long‐term position of the firm in its business environment
to achieve its organizational goals.
• Based on the above, we arrive at a common definition for an e‐Business Strategy.
• An e‐Business strategy is the set of plans and objectives by which applications of internal and external
electronically mediated communication contribute to the corporate strategy.
• Strategic planning comprises a distinct class of decisions (a plan is a set of decisions made for the future) and
objectives, and has to be positioned next to tactical planning (structuring the resources of the firm) and
operational planning (maximizing the profitability of the current operations).
• Strategy is concerned with changes in the competitive environment that may trigger strategic changes for
the individual firm and so affect its roles and functions in the market.
• Reassessment of strategy may occur due to:
o New Products
o Changing customer preferences
Flowers: Roses / Carnations ‐ Orchids
A few years back when people went to the florist, they generally picked up Roses or
Carnations etc. Now, they prefer Orchids. This is an example of changing customer
preferences. A global notion is that a customer does not realize the utility of feel the need
for a product until it is offered to him / her.
o Changing demand patterns
o New competitors
• The frequency, dynamics and predictability of the above changes dictate the intensity of the strategic
planning activity of the firm.
• So, e‐Business strategy (revised) is:
o The set of plans and objectives by which applications of internal and external electronically mediated
communication contribute to the corporate strategy.
• E‐Business strategy may be implemented for:
o Tactical purposes: Mail ‐ EDI ‐XML‐FDI
o Achieving corporate strategy objectives
• E‐Business is strategic in nature.
o The idea is to create a preferably sustainable competitive position for the company.
This is achieved by integration of the Internet and related technologies in its primary
• E‐Business must not only support corporate strategy objectives but also functional strategies (SCM,
• Supply Chain Management Strategy
Vidyalankar Institute of Technology Page 15
o Based on value chain analysis for decomposing an organization into its individual activities and
determining value added at each stage.
o Gauge efficiency in use of resources at each stage.
• Marketing Strategy
o Is a concerned pattern of actions taken in the market environment to create value for the firm by
improving its economic performance.
o Focused on capturing market share or improving profitability via brand‐building etc.
o Operates on CURRENT AS WELL AS FUTURE projections of customer demand.
• Information Systems Strategy
o How to leverage information systems in an organization to support the objectives of an organization
in the long run.
• E‐Business strategy is based on corporate objectives.
Strategic positioning means that a firm is doing things differently from its competitors in a way that delivers a unique
value to its customers. There are 6 fundamental principles a firm must follow to establish and maintain a distinctive
1. Start with the right goal: superior long term ROI.
2. Strategy must enable it to deliver a value proposition different from competitors.
3. Strategy must be reflected in a distinctive value chain.
4. Accept tradeoffs for a robust strategy.
5. Strategy must define how all elements of what a firm does fit together.
6. Strategy must involve continuity of direction.
Levels of eBusiness Strategies
Strategies will exist at different levels of an organization. Strategic levels of management are concerned with
integrating and coordinating the activities of an organization so that the behavior is optimized and its overall
direction is consistent with its mission. Ultimately e‐Business is about communication, within business units and
between units of the enterprise as well as organizations.
1) Supply Chain or Industry Value Chain level
• E‐Business requires a view of the role, added value, and position of the firm in the supply chain.
• Important issues that need to be addressed at this level are:
i. Who are the firm’s direct customers
ii. What is the firm’s value proposal to the customers
iii. Who are the suppliers
iv. How does the firm add value to the suppliers
v. What is the current performance of the Supply Chain in terms of revenue and profitability,
inventory levels etc
vi. More importantly, what are the required performance levels
vii. What are the current problems in the chain
• This sort of analysis give insight into in upstream (supplier side) and downstream (customer side)
data and information flows.
2) The Line of Business or (Strategic) Business Unit level
• Understanding the position in the value chain is a starting point for further analysis of how Internet‐
related technologies could contribute to the competitive strategy of a business.
• This is the level where competitive strategy in a particular market for a particular product is
developed (Strategic Positioning).
Vidyalankar Institute of Technology Page 16
• There are four generic strategies for achieving a profitable business:
i. Differentiation: This strategy refers to all the ways producers can make their product unique
and distinguish them from those of their competitors.
ii. Cost: Adopting a strategy for cost competition means that the company primarily competes
with low cost; customers are interested in buying a product as inexpensively as possible.
Success in such a market implies that the company has discovered a unique business model
which makes it possible to deliver the product or service at the lowest possible cost.
iii. Scope: A scope strategy is a strategy to compete in markets worldwide, rather than merely
in local or regional markets.
iv. Focus: A focus strategy is a strategy to compete within a narrow market segment or product
3) The Corporate or Enterprise level
• This level comprises a collection of (strategic) business units.
• This level addresses the problem of synergy through a firm‐wide, available IT infrastructure.
• Common e‐Business applications throughout the organization are needed for two basic reasons.
• From efficiency point of view, having different applications for the same functionality in different
areas of business is needlessly costly.
• From an effectiveness point of view, there is the need for cross Line of Business communication and
share‐ability of data.
• The emphasis in the business plans is on the customer, not the final product.
• These all become subjects of an enterprise‐wide e‐Business policy.
The changing competitive Agenda: Business Technology Drivers
• Shift in economies from supply driven to demand driven
o Causes a shift in intent of service and quality programs, the impetus for product development the
structure of the organization itself
o One to One marketing
o Mass Customization
o Interactive Nature
o Virtual Nature
Strategic Planning Process
• The strategic planning process has the following steps:
• The strategic planning process starts with the establishment of the organization’s mission statement.
o The mission statement is a basic description of detailing the fundamental purpose of the
organizations existence and encompasses strategy development, including determination of the
organization’s vision and objectives.
o It is developed at the highest level of the organizations management, and provides a general sense
of direction for all decision making within the firm.
• Strategic Analysis
o This involves situation analysis, internal resource assessment, and evaluation of stakeholder’s
Vidyalankar Institute of Technology Page 17
o It will include
Industry or market research
Analysis of Marketplace Structure
Relationships with trading partners and suppliers
Customer Marketing Research
o Information is derived from the analysis of both internal and external factors.
o Internal Factors:
o External Factors:
Legal and regulatory forces
o Any realistic new plan will have to reflect the reality of both the external world and the internal
dynamics of the corporation.
• Strategic Choice
o It is based on the strategic analysis and consists of four parts:
Generation of strategic options
Highlighting possible courses of action
Evaluation of strategic options on their relative merits
Selection of strategy
o Strategic choice results in Strategic Planning, which is concerned with the organizing and detailing of
all the strategies that will be undertaken throughout the organization.
o Planning includes strategy specification and resource allocation.
o It commences with corporate‐level planning that determines the overall direction for the
o This drives Division (or Strategic Business Unit) level planning which deals with groups of related
products offered by the organization.
o These plans in turn become the starting point for operating (or functional) level planning, which
involves more local plans within specific departments of the organization.
o This relates to the actual tasks that must be executed in order to realize a plan and translates
strategy into action.
o It includes monitoring, adjustment, control as well as feedback.
Vidyalankar Institute of Technology Page 18
• In the 1980s the concept of alignment between business and IT was developed.
• According to this concept it is not only feasible to design and build a technically sophisticated infrastructure
for e‐Business, but also to formulate business strategies that complement and support this infrastructure.
• One of the major issues regarding an enterprises investment in IT is whether this is in harmony with its
• This state of harmony is referred to as alignment.
• Alignment is complex, multifaceted and almost never completely achieved. It is about continuing to move in
the right direction and better aligned than the competitors.
• Any e‐Business strategy should articulate an enterprise’s intention to use information technology based on
• When formulating the IT strategy, the enterprise must consider:
o Business objectives and the competitive environment
o Current and future technologies and the costs, risks, and benefits they can bring to the business.
o The capability of the IT organization and technology to deliver current and future levels of service to
o Cost of current IT, and whether this provides sufficient value to the business.
o Lessons learned from past failures and successes.
Consequences of eBusiness
• As e‐Business is an information technology‐enabled organizational phenomenon with economic
consequences, economic theories appear to be particularly useful for analyzing the business effects.
• Strategy is about finding the right (external) fit between organization and environment. Different schools of
thought have approached this problem from different angles.
• When analyzing the business effects of an e‐Business, we will consider the following approaches:
o The Theory of Competitive Strategy
o The resource‐base view
o The theory of transaction costs
• Theory of Competitive Strategy
o The structural attractiveness of a firm is determined by five underlying forces of competition:
The bargaining power of the customers
The bargaining power of the suppliers
The barriers to entry for new competitors
The threat of new substitute products or services
The competition among existing firms in the industry
Vidyalankar Institute of Technology Page 19
o In combination, these forces determine how the economic value created by any product, service
technology or way of competing is divided between companies in an industry.
o The bargaining power of customers for a firm could, for instance, depend on the degree of product
differentiation, and the size of demand and supply. Switching costs are also very important: they
answer the question of how much will it cost the customer to change to another supplier.
o The bargaining power of suppliers is dependent on a variety of factors, such as relative size, number
of suppliers, that can deliver a critical resource, and so on. The Internet causes another specific
threat from the perspective of IT suppliers; they may bypass their customers and directly approach
o The barriers to entry for new competitors depend on how difficult it is to join the industry. Economic
and technological thresholds may prevent outside potential competitors to come in . Economies of
scale, necessary capital, and specialized expertise are important factors in this respect.
o The threat of substitute products depends on the question of whether other products can deliver
added value for consumers instead of current products in the absence of switching costs. e.g. – The
Internet is a serious threat to the Post Office.
o The level of competition among existing firms in the industry will depend on various factors like type
of market, existing competitive behavior, and so on.
• The Resource‐Based View
o According to this theory of economic development, innovation is the source of value creation.
o Several sources of innovation (hence, value creation) are identified:
The introduction of new goods or new production methods,
The creation of new markets,
The discovery of new supply sources,
And the reorganization of industries.
o The resource‐based view (RBV), which builds on the theory of economic development’s perspective
on value creation, regards a firm as a collection of resources and capabilities.
o The RBV looks at available resources first to see how a position in the business environment can be
acquired with them.
o According to this view, a firm can build a strategic position by picking the right resources and
building competencies that are unique and difficult to imitate.
o Resources are considered the raw material for building competencies.
o The RBV states that marshalling and uniquely combing a set of complementary and specialized
resources and capabilities may lead to value creation.
o A firm’s resources and competencies are valuable if, and only if, they reduce a firm’s costs or
improve its revenues.
o Core competencies of an organization encompass knowledge bases, skill sets, and service activities
that can create a continuing competitive advantage.
• Transaction Cost Economics
o Transaction Cost Economics attempt to explain firms’ choices between internalizing and buying
goods and services from the market.
o According to transaction cost theory, exchanges with external firms entail a variety of co‐ordination
costs associated with various aspects of inter‐firm transactions.
o The central question addressed by transaction cost economics is why firms internalize transactions
that might otherwise be conducted in markets. Thus, two key issues concerning firms are:
Which activities should a firm manage within its boundaries, and which activities should it
In which way should a firm manage its relationship with its customers, suppliers and other
Vidyalankar Institute of Technology Page 20
o According to transaction cost economics, a firm has two options for organizing its economic
activities: an internal hierarchical structure where it integrates the activity into its management
structure, or a market‐like relationship with external firms.
o Critical dimensions of transactions influencing the choice of the most effective governance form are:
Specificity of Assets enabling the exchange
o Transaction costs include the costs of planning, adapting, executing and monitoring task completion.
o Transaction cost Theory assumes that markets are not perfect, so lead to costs, like search and
o As internet technology is expected to significantly reduce transaction costs, this theory provides a
basis for assessing the effects of the Internet on new and existing business models.
Success factors for Implementation of eBusiness Strategies
• Transforming an enterprise from a traditional organization to an e‐business based organization’s a complex
• It is essential that senior management develops and endorses a broad strategic vision.
• Once the strategy has been determined and approved the implementation strategy has to be chosen.
• Two approaches prevail:
o The top‐down approach: According to this, business transformation is a business—wide
phenomenon that can only be implemented business wide.
o The bottom‐up approach: In this approach, business re‐engineering starts as an experiment in an
inconspicuous part of an organization. Lessons are learnt from this experiment, and the knowledge is
transferred to other parts of the organization.
• Although the bottom‐up approach has strong support , especially in the case of innovation, central co‐
ordination of the transformation activity is mandatory.
• To provide for the central coordination, program management has to be instituted. A core part of program
management is multi‐project management , the main objectives of which are:
o Recognize dependencies between projects
o Share scarce resources in an overall efficient way
o Systematically utilize experiences from single projects
• Program management is characterized by:
o Program organization,
• Leading a change project or business‐wide initiative requires persons that plan the change and build
business‐wide support; these are called ‘change agents’.
• Change Agents are part of the program management organization.
• In principle, everyone involved in a change project can assume the role of a change agent.
• Three types of change agent roles have been identified:
o Traditional: In the traditional model, the Information System (IS) specialists focus on the delivery of
the implementation of the technology, without considering the organizational aspects. Consequently
they become technicians with a narrow area of expertise.
o Facilitator: In the facilitator model, the central model is that people, not technologies create change.
The change agent brings together all the conditions necessary for the change. In this model, the
change agent remains ‘neutral’, the organization is responsible for the change.
Vidyalankar Institute of Technology Page 21
o Advocate: In this role, change agents focus on inspiring people to adopt the change. Unlike the
facilitator, he does not remain neutral, but uses any tactic (persuasion, manipulation, power etc) to
make the changes accepted.
• Especially in the case of e‐Business transformation, where organizational and IT changes relate to
infrastructure and issues of commonality and interoperability, the advocate model seems to be appropriate.
Pressures Forcing Business Changes
o Fiercer More Global
• Customers have become increasingly demanding
• Integrated Demand (Travel, Car‐Rental etc.)
• Firms ask themselves “which of my competences are unique and of core importance”
o E.g. ‐ Bajaj exits the scooter market.
• Company configuration changes due to outsourcing and in‐sourcing.
There are various definitions for Business Models. The definitions change based on the paradigm and the context
being applied. Let’s look at each definition: ‐
• Participants in a joint business venture:
o Specify the relationships between different participants in a commercial venture, the benefits
costs to each and the flows of revenues. It addresses a simple equation (profit= revenue‐cost)
irrespective of the model.
o Describes how the enterprise produces, delivers and sells its products or services, thus showing how
it delivers value to the customer and how it created wealth.
• Process structure of a business organization:
o Refers to the structures processes in place to operationalize the strategy of business:
o Can be described as:
An architecture for the product, service information flows;
A description of the various business actors their roles;
A description of the potential business benefits for the various actors;
A description of the sources of revenues.
• Perspective of a marketplace.
o Definition can be analyzed from various perspectives:
B2B, B2C activities or both
Position in the value chain
Value proposition target customers
Specific revenue models for generation of it s income streams
Representation Physical or Virtual or combination
• Perspective of e‐Business
o A descriptive representation of the planned activities of an enterprise that involves 3 integral
components which specify: ‐
Internal aspects of a business venture
Type of relationships of the enterprise with its external business environment and its
effective knowledge regarding these relationships
How the information assets are embedded in the business venture.
• A business model can be viewed as an externalization of a firms internal business processes
o Does not involve internal business process complexity.
• When taking the internal aspects of a business into account the following elements need to be defined:
Vidyalankar Institute of Technology Page 22
o Products or Services
o Sources of revenue
o Organization of the firm
E‐Business models are classified as follows:
• Internet Enabled
o Categorized based on increasing functionality, innovation, integration and value.
• Value Web
o Assuredly not a recipe for success but preliminary conceptions of an emerging form of a fluid and
o Move from we‐do‐everything‐ourselves unless (value generated by single organization) to we‐do‐
nothing‐ourselves‐unless (value generated by the network).
• E‐Business Enabled
o Especially valid for B2B contexts
o 5 Representative Business models
o Tele‐working Model:
Collaboration using communication technologies
Classic example is Electronic Manufacturing Services (Solectron)
o Virtual Organization Model:
Collection of geographically dispersed individuals, groups and organizational units.
Example: GeneraL ife (Insurance)
o Process Outsourcing Model:
Example: BPOs, IBM
o Collaborative Product Development Model
Example: Automobile Manufacture ‐ FORD
o Value Chain Integration Model:
Used to improve communication collaboration between all supply chain parties.
• Market Participants
o More generic classification of Internet Based Business Models
o In disagreement with the widely accepted idea that e‐Business will cause industry value chains to be
restructured to such an extent that intermediation will no longer be a prominent feature.
o The real trend might just be towards an increase in intermediation by cyber‐mediaries.
Organizations which operate in electronic markets to facilitate exchanges between
producers and consumers by meeting the needs of both.
Directories of Directory Services Intermediaries
Spot Market Makers
Financial Intermediaries (ESCROW Service for Online purchases).
Note: For details on all e‐Business models, Refer Michael Papazoglou, “e‐Business – Organizational Technical
Vidyalankar Institute of Technology Page 23
Integration of Applications
eBusiness Integration (Patterns)
e‐Business Integration occurs in as many forms as there are e‐Businesses. At first glance, integration problems and
the corresponding solutions are seldom identical. Yet, upon closer examination, you discover that integration
solutions can actually be classified into common categories. Each of these categories describes both a "type" of
integration problem as well as a solution method. These categories are called integration patterns. Integration
patterns help you understand the different methods available to you for a given type of integration problem. They
allow you to take a step back and understand the differences in the various scenarios and appreciate the different
approaches to integration. Finally, they allow you to view "integration in the big picture." You can learn to break
down what may be a complex integration into conceptual categories and understand which technologies to apply.
What Are Integration Patterns
A pattern is commonly defined as a reliable sample of traits, acts, tendencies, or other observable characteristics. In
software development, you may be familiar with the idea of design patterns or process patterns. Design patterns
systematically describe object designs that can be employed for a common set of problems. Similarly, process
patterns describe proven methods and processes used in software development. In practice, patterns are simply a
logical classification of commonly recurring actions, techniques, designs, or organizations. What are integration
patterns Integration patterns emerge from classification of standard solutions for integration scenarios. They are
not patterns of design or code. Nor are they patterns of operational processes for an integration project. Instead,
each integration pattern defines a type of integration problem, a solution technique, as well as parameters applied
for e‐Business Integration.
Following are seven common e‐Business Integration patterns. They are not meant to be comprehensive, but they
cover most of the common integration scenarios implemented today. They encompass both EAI scenarios as well as
• EAI (intra‐enterprise) Patterns
1. Database Replication
2. Single‐Step Application Integration
3. Multi‐Step Application Integration
4. Brokering Application
• B2Bi (inter‐enterprise) Patterns
5. Application‐to‐Application B2Bi
6. Data Exchange B2Bi
7. B2B Process Integration
The EAI Patterns represent patterns commonly applied within a corporate enterprise, whereas the B2Bi Patterns
represent the different methods in conducting integrated B2B transactions. The following sections provide a closer
look at each of these patterns and discuss some of the details.
The Database Replication pattern may be the most prevalent pattern of EAI integration today. Database replication
involves managing copies of data over two or more databases, resulting in redundant data. Companies engage in
database replication for numerous reasons. One reason is that many organizations are becoming more distributed in
their operations, requiring multiple copies of the same data over several physical locations. Replication is also a
means of data recovery. In many organizations, an active secondary database is maintained for data recovery
purposes. In the event that the production database needs to be recovered, the secondary replicated database can
be used. This also applies for "high availability" systems. In these situations, a redundant copy of "live" data is
Vidyalankar Institute of Technology Page 24
maintained to ensure that if the first system is not available, the redundant database system is activated. The two
general categories for database replication are synchronous and asynchronous replication.
SingleStep Application Integration
The Single‐Step Application Integration (SSAI) pattern extends the asynchronous database replication pattern.
Instead of focusing on data consistency between two databases, the SSAI pattern integrates data between
applications, moving data from one context to another. It does so by translating data syntax of the source message
and reformatting data elements into a new target message. It is "single step" because it requires an intermediary
broker to map source messages to target messages. Typically, it is an extension of the asynchronous replication
technology, in that it utilizes Message Queuing Middleware such as MQ Series. It is just as likely to be implemented
with the less sophisticated FTP in a batch mode. In either case, the point is that it does more than simply move data
from point A to point B for consistency’s sake. Whereas, in the replication pattern both the source and target data
models are likely similar, if not identical at times, this is not necessarily the case for the SSAI pattern. The objective
here is not data consistency, but application data integration.
MultiStep Application Integration
The Multi‐Step Application Integration (MSAI) pattern is an extension of the SSAI pattern. MSAI enables the
integration of n (source) to m (target) applications. It addresses many‐to‐many integration, which SSAI cannot, by
providing what is known as sequential logical processing. In other words, steps in this pattern are processed
sequentially, and rules applied are Boolean logical in nature. Like the single‐step pattern, MSAI requires an
intermediary to broker the transaction of data between applications. It is often built around an asynchronous event‐
based system and typically is implemented through the use of Message Queuing Middleware as well. The
asynchronous eventbased approach creates loose coupling. Although each system is physically independent, they
are logically dependent. In other words, interdependencies exist between the application events that can be
expressed in terms of transformations and data integration rules. Data elements from one application can drive the
retrieval or processing of messages in another application. The simplest multi‐step example in Figure 3.3 involves
three applications in which a message from application A is combined with a message from application B that is
reformatted for a target application C. It is common for a data element from application A to act as a key to drive the
request for information from application B.
At times integrating two applications is not principally a matter of integrating data, but integrating business logic.
The Brokering Application pattern addresses the use of intermediary application logic to link together two or more
applications. In plain terms, it means that custom application code is written containing logic to broker interactions
between the disparate applications. This custom brokering application sits in the middle as an intermediary for
processing requests from different applications
The use of this solution pattern is particularly applicable in the scenarios below:
• Applications Need to Reuse Logic
• Applications Linked by Complex Logic
• Applications Unified Through User Interface
Now you're ready to move beyond EAI to learn about Application‐to‐Application B2Bi, extending integration beyond
the corporate enterprise. I will describe four additional patterns related specifically to B2B integration, beginning
first with the Application‐to‐Application B2Bi pattern. The Application‐to‐Application pattern is the logical extension
of what occurs in EAI. When EAI vendors tout their products as being B2Bi, this specific pattern is what they have in
mind. However, as you will discover, this is not the only pattern and very likely not even the primary pattern for
B2Bi. Application‐to‐Application B2Bi, which is often referred to as inter‐enterprise integration, involves corporate
Vidyalankar Institute of Technology Page 25
entities linking their applications directly to the applications of their partners or customers. In practice, this type of
integration is often implemented as part of a supply chain of goods and services to the customer.
This extension for interenterprise integration means that a number of additional issues need to be accounted for:
• Federated Control
• Systems Management
Data Exchange B2Bi
The limitation of the Application‐to‐Application B2Bi pattern is that it can be more demanding to implement. It
necessitates that each participant handles and externalizes application native data directly. This makes it difficult to
scale the B2B interaction model rapidly when such a demand is placed on the participants. The optimal solution is to
provide a rapidly scalable B2Bi model in which participants can exchange data freely with minimal expectation on
their infrastructure. The Data Exchange B2Bi pattern enables B2B transactions predicated on a common data
exchange format. It is the most widely applied pattern for B2B commerce today. Data Exchange B2Bi is effective
because it is simple in concept and has been in use since the days of Electronic Data Interchange (EDI), the
forerunner to today’s B2B over the Internet.
Although there is a significant incumbency of legacy EDI transactions, the XML‐based B2B will ultimately displace EDI
as the primary mechanism for e‐Business transactions. XML‐based data packets are transmitted between two
business entities through the use of a data exchange gateway service on both ends. One of the primary
responsibilities of the gateway service is to prepare the data packets by placing them within a security envelope. The
B2B gateway service supports security standards such as MIME, X.509, and S/Key. It is also responsible for routing
data through a standard transport. Most B2B gateway services provide numerous transport options including HTTPS,
FTP, and TCP/IP Sockets. However, upon examination, you will find that most B2Bi transactions still deliver XML
documents over an HTTPS pipe.
B2B Process Integration
Even with industrywide initiatives such as RosettaNet, a point‐to‐point data exchange that manages static
interactions has some limitations. If Corporation A wants to purchase office supplies from Depot X, it must agree
ahead of time on the content of the documents exchanged and buying process. This is, of course, to be expected.
However, what if the situation involves managing multiple suppliers or if the interactions become more complex
For instance, a scenario in which suppliers openly bid to compete on pricing will increase the dimensions of process
interactions. In that case, managing the B2B transaction is no longer an activity of managing a single point‐to‐point
interaction. Instead, it becomes a challenge of managing business processes that are dynamic rather than static.
The B2B Process Integration pattern takes the limitations raised by the Data Exchange pattern and addresses them
by providing Business Process Integration (BPI) services. Just as the Data Exchange pattern allows participants to
manage data exchanges dynamically through XML‐based documents, the B2B Process Integration pattern allows the
participants to manage processes in the same way.
Therefore, richer, more complex relationships can occur between trading partners. B2B Process Integration pattern
can be implemented as one of two variations: Closed Process B2Bi or Open Process B2Bi. You might argue that each
of these variations constitutes an individual pattern, but because they share the common attribute of being process
focused, I have decided to treat them as variations to the B2B Process Integration pattern.
Approaches to Middleware
Middleware is computer software that connects software components or some people and their applications. The
software consists of a set of services that allows multiple processes running on one or more machines to interact.
This technology evolved to provide for interoperability in support of the move to coherent distributed architectures,
Vidyalankar Institute of Technology Page 26
which are most often used to support and simplify complex distributed applications. It includes web servers,
application servers, and similar tools that support application development and delivery. Middleware is especially
integral to modern information technology based on XML, SOAP, Web services, and service‐oriented architecture.
Middleware sits "in the middle" between application software that may be working on different operating systems.
It is similar to the middle layer of a three‐tier single system architecture, except that it is stretched across multiple
systems or applications. Examples include EAI software, telecommunications software, transaction monitors, and
The distinction between operating system and middleware functionality is, to some extent, arbitrary. While core
kernel functionality can only be provided by the operating system itself, some functionality previously provided by
separately sold middleware is now integrated in operating systems. A typical example is the TCP/IP stack for
telecommunications, nowadays included in virtually every operating system.
In simulation technology, middleware is generally used in the context of the high level architecture (HLA) that applies
to many distributed simulations. It is a layer of software that lies between the application code and the run‐time
infrastructure. Middleware generally consists of a library of functions, and enables a number of applications—
simulations or federates in HLA terminology—to page these functions from the common library rather than re‐create
them for each application
Definition of Middleware
Software that provides a link between separate software applications. Middleware is sometimes called plumbing
because it connects two applications and passes data between them. Middleware allows data contained in one
database to be accessed through another. This definition would fit enterprise application integration and data
ObjectWeb defines middleware as: "The software layer that lies between the operating system and applications on
each side of a distributed computing system in a network."
Middleware is computer software that connects software components or applications. The software consists of a set
of services that allows multiple processes running on one or more machines to interact. This technology evolved to
provide for interoperability in support of the move to coherent distributed architectures, which are most often used
to support and simplify complex, distributed applications.
It includes web servers, application servers, and similar tools that support application development and delivery.
Middleware is especially integral to modern information technology based on XML, SOAP, Web services, and service‐
In simulation technology, middleware is generally used in the context of the high level architecture (HLA) that applies
to many distributed simulations. It is a layer of software that lies between the application code and the run‐time
infrastructure. Middleware generally consists of a library of functions, and enables a number of applications—
simulations or federates in HLA terminology—to page these functions from the common library rather than re‐create
them for each application.
Origin of Middleware
Middleware is a relatively new addition to the computing landscape. It gained popularity in the 1980s as a solution
to the problem of how to link newer applications to older legacy systems, although the term had been in use since
1968. It also facilitated distributed processing, the connection of multiple applications to create a larger application,
usually over a network.
Vidyalankar Institute of Technology Page 27
Use of middleware
Middleware services provide a more functional set of application programming interfaces to allow an application to
(when compared to the operating system and network services.):
• Locate transparently across the network, thus providing interaction with another service or application
• Filter data to make them friendly usable or public via anonymization process for privacy protection (for
• Be independent from network services
• Be reliable and always available
• Add complementary attributes like semantics
Middleware offers some unique technological advantages for business and industry. For example, traditional
database systems are usually deployed in closed environments where users access the system only via a restricted
network or intranet (e.g., an enterprise's internal network). With the phenomenal growth of the World Wide Web,
users can access virtually any database for which they have proper access rights from anywhere in the world.
Middleware addresses the problem of varying levels of interoperability among different database structures.
Middleware facilitates transparent access to legacy database management systems (DBMSs) or applications via a
web server without regard to database‐specific characteristics .
Businesses frequently use middleware applications to link information from departmental databases, such as payroll,
sales, and accounting, or databases housed in multiple geographic locations. In the highly competitive healthcare
community, laboratories make extensive use of middleware applications for data mining, laboratory information
system (LIS) backup, and to combine systems during hospital mergers. Middleware helps bridge the gap between
separate LISs in a newly formed healthcare network following a hospital buyout.
Wireless networking developers can use middleware to meet the challenges associated with wireless sensor network
(WSN), or WSN technologies. Implementing a middleware application allows WSN developers to integrate operating
systems and hardware with the wide variety of various applications that are currently available.
Middleware can help software developers avoid having to write application programming interfaces (API) for every
control program, by serving as an independent programming interface for their applications. For Future Internet
network operation through traffic monitoring in multi‐domain scenarios, using mediator tools (middleware) is a
powerful help since they allow operators, searchers and service providers to supervise Quality of service and analyse
eventual failures in telecommunication services.
Finally, e‐commerce uses middleware to assist in handling rapid and secure transactions over many different types of
computer environments. In short, middleware has become a critical element across a broad range of industries,
thanks to its ability to bring together resources across dissimilar networks or computing platforms.
Types of middleware
Hurwitz's classification system organizes the many types of middleware that are currently available. These
classifications are based on scalability and recoverability:
• Remote Procedure Call — Client makes calls to procedures running on remote systems. Can be
asynchronous or synchronous.
• Message Oriented Middleware — Messages sent to the client are collected and stored until they are
acted upon, while the client continues with other processing.
• Object Request Broker — This type of middleware makes it possible for applications to send objects and
request services in an object‐oriented system.
• SQL‐oriented Data Access — middleware between applications and database servers.
Vidyalankar Institute of Technology Page 28
• Embedded Middleware — communication services and integration interface software/firmware that
operates between embedded applications and the real time operating system.
• Other sources include these additional classifications:
• Transaction processing monitors — Provides tools and an environment to develop and deploy
• Application servers — software installed on a computer to facilitate the serving (running) of other
• Enterprise Service Bus — An abstraction layer on top of an Enterprise Messaging System.
In computer science, a remote procedure call (RPC) is an inter‐process communication that allows a computer
program to cause a subroutine or procedure to execute in another address space (commonly on another computer
on a shared network) without the programmer explicitly coding the details for this remote interaction. That is, the
programmer writes essentially the same code whether the subroutine is local to the executing program, or remote.
When the software in question uses object‐oriented principles, RPC is called remote invocation or remote method
invocation. Note that there are many different (often incompatible) technologies commonly used to accomplish this.
History and origins
The idea of RPC (Remote Procedure Call) goes back at least as far as 1976, when it was described in RFC 707. One of
the first business uses of RPC was by Xerox under the name "Courier" in 1981. The first popular implementation of
RPC on Unix was Sun's RPC (now called ONC RPC), used as the basis for NFS (Sun). Another early Unix
implementation was Apollo Computer's Network Computing System (NCS). NCS later was used as the foundation of
DCE/RPC in the OSF's Distributed Computing Environment (DCE). A decade later Microsoft adopted DCE/RPC as the
basis of the Microsoft RPC (MSRPC) mechanism, and implemented DCOM on top of it. Around the same time (mid‐
90's), Xerox PARC's ILU, and the Object Management Group's CORBA, offered another RPC paradigm based on
distributed objects with an inheritance mechanism.
An RPC is initiated by the client, which sends a request message to a known remote server to execute a specified
procedure with supplied parameters. The remote server sends a response to the client, and the application
continues its process. There are many variations and subtleties in various implementations, resulting in a variety of
different (incompatible) RPC protocols. While the server is processing the call, the client is blocked (it waits until the
server has finished processing before resuming execution). An important difference between remote procedure calls
and local calls is that remote calls can fail because of unpredictable network problems. Also, callers generally must
deal with such failures without knowing whether the remote procedure was actually invoked. Idempotent
procedures (those that have no additional effects if called more than once) are easily handled, but enough
difficulties remain that code to call remote procedures is often confined to carefully written low‐level subsystems.
The steps in making a RPC
1. The client calling the Client stub. The call is a local procedure call, with parameters pushed on to the stack in
the normal way.
2. The client stub packing the parameters into a message and making a system call to send the message.
Packing the parameters is called marshaling.
3. The kernel sending the message from the client machine to the server machine.
4. The kernel passing the incoming packets to the server stub.
5. Finally, the server stub calling the server procedure. The reply traces the same in other direction.
Vidyalankar Institute of Technology Page 29
Standard contact mechanisms
To let different clients access servers, a number of standardized RPC systems have been created. Most of these use
an interface description language (IDL) to let various platforms call the RPC. The IDL files can then be used to
generate code to interface between the client and server. The most common tool used for this is RPCGEN.
Other RPC analogues
RPC analogues found elsewhere:
• Java's Java Remote Method Invocation (Java RMI) API provides similar functionality to standard UNIX
• Modula‐3's Network Objects, which were the basis for Java's RMI
• XML‐RPC is an RPC protocol that uses XML to encode its calls and HTTP as a transport mechanism.
• Microsoft .NET Remoting offers RPC facilities for distributed systems implemented on the Windows
• RPyC implements RPC mechanisms in Python, with support for asynchronous calls.
• Pyro Object Oriented form of RPC for Python.
• Etch (protocol) framework for building network services.
• Facebook's Thrift protocol and framework.
• CORBA provides remote procedure invocation through an intermediate layer called the "Object Request
• DRb allows Ruby programs to communicate with each other on the same machine or over a network.
• remote method invocation (RMI) to pass commands and data between processes.
• AMF allows Flex applications to communicate with back‐ends or other applications that support AMF.
• Libevent provides a framework for creating RPC servers and clients.
• Windows Communication Foundation is an application
The Java Remote Method Invocation Application Programming Interface (API), or Java RMI, is a Java application
programming interface that performs the object‐oriented equivalent of remote procedure calls (RPC).
1. The original implementation depends on Java Virtual Machine (JVM) class representation mechanisms and it
thus only supports making calls from one JVM to another. The protocol underlying this Java‐only
implementation is known as Java Remote Method Protocol (JRMP).
2. In order to support code running in a non‐JVM context, a CORBA version was later developed.
Vidyalankar Institute of Technology Page 30
Usage of the term RMI may denote solely the programming interface or may signify both the API and JRMP, whereas
the term RMI‐IIOP (read: RMI over IIOP) denotes the RMI interface delegating most of the functionality to the
supporting CORBA implementation.
The programmers of the original RMI API generalized the code somewhat to support different implementations,
such as a HTTP transport. Additionally, the ability to pass arguments "by value" was added to CORBA in order to
support the RMI interface. Still, the RMI‐IIOP and JRMP implementations do not have fully identical interfaces.
RMI functionality comes in the package java.rmi, while most of Sun's implementation is located in the sun.rmi
package. Note that with Java versions before Java 5.0 developers had to compile RMI stubs in a separate compilation
step using rmic. Version 5.0 of Java and beyond no longer require this step.
Jini offers a more advanced version of RMI in Java. It functions similarly but provides more advanced searching
capabilities and mechanisms for distributed object applications.
The following classes implement a simple client‐server program using RMI that displays a message. RmiServer
class—Listens to RMI requests and implements the interface which is used by the client to invoke remote
public class RmiServer extends UnicastRemoteObject
static public final String MESSAGE = "Hello world";
public RmiServer() throws RemoteException
public String getMessage()
public static void main(String args)
System.out.println("RMI server started");
// Create and install a security manager
if (System.getSecurityManager() == null)
System.out.println("Security manager installed.");
System.out.println("Security manager already exists.");
try //special exception handler for registry creation
System.out.println("java RMI registry created.");
catch (RemoteException e)
//do nothing, error means registry already exists
System.out.println("java RMI registry already exists.");
Vidyalankar Institute of Technology Page 31
RmiServer obj = new RmiServer();
// Bind this object instance to the name "RmiServer"
System.out.println("PeerServer bound in registry");
catch (Exception e)
System.err.println("RMI server exception:");
RmiServerIntf class—Defines the interface that is used by the client and implemented by the server.
public interface RmiServerIntf extends Remote
public String getMessage() throws RemoteException;
RmiClient class—This is the client which gets the reference to the remote object and invokes its method to get a
public class RmiClient
// "obj" is the reference of the remote object
RmiServerIntf obj = null;
public String getMessage()
catch (Exception e)
System.err.println("RmiClient exception: " +
public static void main(String args)
// Create and install a security manager
if (System.getSecurityManager() == null)
RmiClient cli = new RmiClient();
Before running this subj, we need to make 'Stub' file of interface we used. For this task we have RMI compiller ‐
• Note: we make stub file from .class with implementation remote interface, not '.java'
Vidyalankar Institute of Technology Page 32
server.policy—This file is required on the server to allow TCP/IP communication for the remote registry and for
the RMI server.
permission java.net.SocketPermission "127.0.0.1:",
permission java.net.SocketPermission "127.0.0.1:", "accept";
The server.policy file should be used using the D switch of Java RTE, e.g.:
java.exe Djava.security.policy=server.policy RmiServer
client.policy—This file is required on the client to connect to RMI Server using TCP/IP.
permission java.net.SocketPermission "127.0.0.1:",
no.policy—Also if you have a troubles with connecting, try this file for server or client.
Vidyalankar Institute of Technology Page 33
Enterprise Application Integration
Enterprise Application Integration (EAI) is defined as the use of software and computer systems architectural
principles to integrate a set of enterprise computer applications. Enterprise Application Integration (EAI) is an
integration framework composed of a collection of technologies and services which form a middleware to enable
integration of systems and applications across the enterprise. Supply chain management applications (for managing
inventory and shipping), customer relationship management applications (for managing current and potential
customers), business intelligence applications (for finding patterns from existing data from operations), and other
types of applications (for managing data such as human resources data, health care, internal communications, etc)
typically cannot communicate with one another in order to share data or business rules.
For this reason, such applications are sometimes referred to as islands of automation or information silos. This lack
of communication leads to inefficiencies, wherein identical data are stored in multiple locations, or straightforward
processes are unable to be automated. Enterprise application integration (EAI) is the process of linking such
applications within a single organization together in order to simplify and automate business processes to the
greatest extent possible, while at the same time avoiding having to make sweeping changes to the existing
applications or data structures. In the words of the Gartner Group, EAI is the “unrestricted sharing of data and
business processes among any connected application or data sources in the enterprise.”
One large challenge of EAI is that the various systems that need to be linked together often reside on different
operating systems, use different database solutions and different computer languages, and in some cases are legacy
systems that are no longer supported by the vendor who originally created them. In some cases, such systems are
dubbed "stovepipe systems" because they consist of components that have been jammed together in a way that
makes it very hard to modify them in any way.
Purposes of EAI
EAI can be used for different purposes:
• Data (information) Integration: Ensuring that information in multiple systems is kept consistent. This is
also known as EII (Enterprise Information Integration).
• Vendor independence: Extracting business policies or rules from applications and implementing them in
the EAI system, so that even if one of the business applications is replaced with a different vendor's
application, the business rules do not have to be re‐implemented.
• Common Facade: An EAI system could front‐end a cluster of applications, providing a single consistent
access interface to these applications and shielding users from having to learn to interact with different
There are two patterns that EAI systems implement:
• Mediation: Here, the EAI system acts as the go‐between or broker between (interface or
communicating) multiple applications. Whenever an interesting event occurs in an application (e. g., new
information created, new transaction completed, etc.) an integration module in the EAI system is
notified. The module then propagates the changes to other relevant applications.
• Federation: In this case, the EAI system acts as the overarching facade across multiple applications. All
event calls from the 'outside world' to any of the applications are front‐ended by the EAI system. The EAI
system is configured to expose only the relevant information and interfaces of the underlying
applications to the outside world, and performs all interactions with the underlying applications on
behalf of the requester.
Vidyalankar Institute of Technology Page 34
Both patterns are often used concurrently. The same EAI system could be keeping multiple applications in sync
(mediation), while servicing requests from external users against these applications (federation).
EAI supports both asynchronous and synchronous access patterns, the former being typical in the mediation case
and the latter in the federation case.
An integration operation could be short‐lived (e. g., keeping data in sync across two applications could be completed
within a second) or long‐lived (e. g., one of the steps could involve the EAI system interacting with a human work
flow application for approval of a loan that takes hours or days to complete).
There are two major topologies: hub‐and‐spoke, and bus. Each has its own advantages and disadvantages. In the
hub‐and‐spoke model, the EAI system is at the center (the hub), and interacts with the applications via the spokes. In
the bus model, the EAI system is the bus (or is implemented as a resident module in an already existing message bus
or message‐oriented middleware).
Multiple technologies are used in implementing each of the components of the EAI system:
This is usually implemented by enhancing standard middleware products (application server, message bus)
or implemented as a stand‐alone program (i. e., does not use any middleware), acting as its own
The bus/hub connects to applications through a set of adapters (also referred to as connectors). These are
programs that know how to interact with an underlying business application. The adapter performs two‐way
communication, performing requests from the hub against the application, and notifying the hub when an
event of interest occurs in the application (a new record inserted, a transaction completed, etc.). Adapters
can be specific to an application (e. g., built against the application vendor's client libraries) or specific to a
class of applications (e. g., can interact with any application through a standard communication protocol,
such as SOAP or SMTP). The adapter could reside in the same process space as the bus/hub or execute in a
remote location and interact with the hub/bus through industry standard protocols such as message queues,
web services, or even use a proprietary protocol. In the Java world, standards such as JCA allow adapters to
be created in a vendor‐neutral manner.
Data format and transformation
To avoid every adapter having to convert data to/from every other applications' formats, EAI systems usually
stipulate an application‐independent (or common) data format. The EAI system usually provides a data
transformation service as well to help convert between application‐specific and common formats. This is
done in two steps: the adapter converts information from the application's format to the bus's common
format. Then, semantic transformations are applied on this (converting zip codes to city names,
splitting/merging objects from one application into objects in the other applications, and so on).
An EAI system could be participating in multiple concurrent integration operations at any given time, each
type of integration being processed by a different integration module. Integration modules subscribe to
Vidyalankar Institute of Technology Page 35
events of specific types and process notifications that they receive when these events occur. These modules
could be implemented in different ways: on Java‐based EAI systems, these could be web applications or EJBs
or even POJOs that conform to the EAI system's specifications.
Support for transactions
When used for process integration, the EAI system also provides transactional consistency across
applications by executing all integration operations across all applications in a single overarching distributed
transaction (using two‐phase commit protocols or compensating transactions).
Currently, there are many variations of thought on what constitutes the best infrastructure, component model, and
standards structure for Enterprise Application Integration. There seems to be consensus that four components are
essential for modern enterprise application integration architecture:
1. A centralized broker that handles security, access, and communication. This can be accomplished through
integration servers (like the School Interoperability Framework (SIF) Zone Integration Servers) or through
similar software like the Enterprise service bus (ESB) model that acts as a SOAP‐oriented services manager.
2. An independent data model based on a standard data structure, also known as a Canonical data model. It
appears that XML and the use of XML style sheets has become the de facto and in some cases de jure
standard for this uniform business language.
3. A connector or agent model where each vendor, application, or interface can build a single component that
can speak natively to that application and communicate with the centralized broker.
4. A system model that defines the APIs, data flow and rules of engagement to the system such that
components can be built to interface with it in a standardized way.
Although other approaches like connecting at the database or user‐interface level have been explored, they have not
been found to scale or be able to adjust. Individual applications can publish messages to the centralized broker and
subscribe to receive certain messages from that broker. Each application only requires one connection to the broker.
This central control approach can be extremely scalable and highly evolvable. Enterprise Application Integration is
related to middleware technologies such as message‐oriented middleware (MOM), and data representation
technologies such as XML. Other EAI technologies involve using web services as part of service‐oriented architecture
as a means of integration. Enterprise Application Integration tends to be data centric. In the near future, it will come
to include content integration and business processes.
EAI Implementation Pitfalls
In 2003 it was reported that 70 of all EAI projects fail.
Most of these failures are not due to the software itself or technical difficulties, but due to management issues.
Integration Consortium European Chairman Steve Craggs has outlined the seven main pitfalls undertaken by
companies using EAI systems and explains solutions to these problems.
• Constant change
The very nature of EAI is dynamic and requires dynamic project managers to manage their implementation.
• Shortage of EAI experts
EAI requires knowledge of many issues and technical aspects.
• Competing standards
Within the EAI field, the paradox is that EAI standards themselves are not universal.
Vidyalankar Institute of Technology Page 36
• EAI is a tool paradigm
EAI is not a tool, but rather a system and should be implemented as such.
• Building interfaces is an art
Engineering the solution is not sufficient. Solutions need to be negotiated with user departments to reach a
common consensus on the final outcome. A lack of consensus on interface designs leads to excessive effort
to map between various systems data requirements.
• Loss of detail
Information that seemed unimportant at an earlier stage may become crucial later.
Since so many departments have many conflicting requirements, there should be clear accountability for the
system's final structure.
Other potential problems may arise in these areas:
• Emerging Requirements
EAI implementations should be extensible and modular to allow for future changes.
The applications whose data is being integrated often belong to different departments that have technical,
cultural, and political reasons for not wanting to share their data with other departments
Advantages and Disadvantages
o Real time information access among systems
o Streamlines business processes and helps raise organizational efficiency
o Maintains information integrity across multiple systems
o Ease of development and maintenance
o High initial development costs, especially for small and mid‐sized businesses (SMBs)
o Require a fair amount of up front business design, which many managers are not able to envision or
not willing to invest in. Most EAI projects usually start off as point‐to‐point efforts, very soon
becoming unmanageable as the number of applications increase
Software as a Service (SaaS)
Software as a Service is a relatively new trend among software providers in which an application is designed to be
accessed on the Web rather than installed on the user’s workstation. Though the software is Web‐based, it is not
browser‐based. Instead, the software provider develops a thin client application that the user can download one
time to their workstation. Once the application is installed, the user’s workstation maintains constant
communication with the software provider’s server whenever the application is in use. The application is called a
thin client because it provides a considerably smaller download than locally housed applications, or fat clients, while
maintaining the same robust functionality. The hosted server houses all of the user’s data; none is stored on the
user’s computer. There are many benefits to using hosted software.
Vidyalankar Institute of Technology Page 37
Service‐oriented architecture (SOA) is a hot topic in enterprise computing because many IT professionals see the
potential of an SOA ‐‐ especially a web services‐based SOA ‐‐ in dramatically speeding up the application
development process. They also see it as a way to build applications and systems that are more adaptable, and in
doing so, they see IT becoming more agile in responding to changing business needs. Not only is SOA a hot topic, but
it's clearly the wave of the future. Gartner reports that "By 2008, SOA will be a prevailing software engineering
practice, ending the 40‐year domination of monolithic software architecture" and that "Through 2008, SOA and web
services will be implemented together in more than 75 percent of new SOA or web services projects."
But despite this strong trend, some in the IT community don't feel that the web services underpinning for an SOA is
mature enough for their enterprise to consider migration to a service‐oriented architecture. For others, the terms
service‐oriented architecture and web services draw a blank stare. An earlier article, presented a brief overview of
SOA and the role of web services in realizing it. This article supplements that earlier article. If you're not familiar with
SOA and web services, this article aims to familiarize you with them. It defines some of the key terms and concepts
related to SOA and web services. A critical mass of widely‐adopted technologies is available now to implement and
use a web services‐based SOA, and more technologies, as well as tools, are on the way. Figure 1 identifies some of
these technologies and tools. Each layer of the figure shows technologies or tools that leverage technologies in the
A service‐oriented architecture is an information technology approach or strategy in which applications make use of
(perhaps more accurately, rely on) services available in a network such as the World Wide Web. Implementing a
service‐oriented architecture can involve developing applications that use services, making applications available as
services so that other applications can use those services, or both. A service provides a specific function, typically a
business function, such as analyzing an individual's credit history or processing a purchase order. A service can
provide a single discrete function, such as converting one type of currency into another, or it can perform a set of
related business functions, such as handling the various operations in an airline reservations system. Services that
perform a related set of business functions, as opposed to a single function, are said to be "coarse grained." Multiple
services can be used together in a coordinated way. The aggregated, or composite, service can be used to satisfy a
more complex business requirement. In fact, one way of looking at an SOA is as an approach to connecting
applications (exposed as services) so that they can communicate with (and take advantage of) each other. In other
words, a serviceoriented architecture is a way of sharing functions (typically business functions) in a widespread and
Vidyalankar Institute of Technology Page 38
The concept of an SOA is not new. Service‐oriented architectures have been used for years. What distinguishes an
SOA from other architectures is loose coupling. Loose coupling means that the client of a service is essentially
independent of the service. The way a client (which can be another service) communicates with the service doesn't
depend on the implementation of the service. Significantly, this means that the client doesn't have to know very
much about the service to use it. For instance, the client doesn't need to know what language the service is coded in
or what platform the service runs on. The client communicates with the service according to a specified, well‐defined
interface, and then leaves it up to the service implementation to perform the necessary processing. If the
implementation of the service changes, for instance, the airline reservations application is revised, the client
communicates with it in the same way as before, provided that the interface remains the same. Loose coupling
enables services to be documentoriented (or document‐centric). A document‐oriented service accepts a document
as input, as opposed to something more granular like a numeric value or Java object. The client does not know or
care what business function in the service will process the document. It's up to the service to determine what
business function (or functions) to apply based on the content of the document. However what is relatively new is
the emergence of web services‐based SOAs. A web service is a service that communicates with clients through a set
of standard protocols and technologies. These web services standards are implemented in platforms and products
from all the major software vendors, making it possible for clients and services to communicate in a consistent way
across a wide spectrum of platforms and operating environments. This universality has made web services the most
prevalent approach to implementing an SOA.
Optionally, an SOA can also include a service that provides a directory or registry of services. The registry contains
information about the service such as it's interface. A client can discover services by examining the registry. A
registry can also be coupled with a repository component that stores additional information about each service. This
additional "metadata" can include business process information such as policy statements.
There are many reasons for an enterprise to take an SOA approach, and more specifically, a web servicesbased SOA
approach. Some of the primary reasons are:
Reusability: What drives the move to SOA is reuse of business services. Developers within an enterprise and across
enterprises (particularly, in business partnerships) can take the code developed for existing business applications,
expose it as web services, and then reuse it to meet new business requirements. Reusing functionality that already
exists outside or inside an enterprise instead of developing code that reproduces those functions can result in a huge
savings in application development cost and time. The benefit of reuse grows dramatically as more and more
business services get built, and incorporated into different applications. A major obstacle in taking advantage of
existing code is the uniqueness of specific applications and systems. Typically, solutions developed in different
enterprises, even different departments within the same enterprise, have unique characteristics. They run in
different operating environments, they're coded in different languages, they use different programming interfaces
and protocols. You need to understand how and where these applications and systems run to communicate with
them. The work involved in doing this analysis and the development effort in tying these pieces together can be very
time consuming. Witness the pain IT organizations generally encounter when they try to integrate their applications
with systems from business partners (or even with legacy systems from other parts of their own company). In an
SOA, the only characteristic of a service that a requesting application needs to know about is the public interface.
The functions of an application or system (including legacy systems) can be dramatically easier to access as a service
in an SOA than in some other architecture. So integrating applications and systems can be much simpler.
Interoperability: The SOA vision of interaction between clients and loosely‐coupled services means widespread
interoperability. In other words, the objective is for clients and services to communicate and understand each other
no matter what platform they run on. This objective can be met only if clients and services have a standard way of
communicating with each other ‐‐ a way that's consistent across platforms, systems, and languages. In fact, web
services provide exactly that. Web services comprise a maturing set of protocols and technologies that are widely
Vidyalankar Institute of Technology Page 39
accepted and used, and that are platform, system, and language independent. In addition, these protocols and
technologies work across firewalls, making it easier for business partners to share vital services. Promising to make
things even more consistent is the WS‐I basic profile, introduced by the Web Services Interoperability Organization
(an organization chartered to promote web services interoperability). The WS‐I basic profile identifies a core set of
web services technologies that when implemented in different platforms and systems, helps ensure that services on
these different platforms and systems, and written in different languages, can communicate with each other. The
WS‐I basic profile has widespread backing in the computer industry, virtually guaranteeing interoperability of
services that conform to the profile.
Scalability: Because services in an SOA are loosely coupled, applications that use these services tend to scale easily ‐‐
certainly more easily than applications in a more tightly‐coupled environment. That's because there are few
dependencies between the requesting application and the services it uses. The dependencies between client and
service in a tightly‐coupled environment are compounded (and the development effort made significantly more
complex) as an application that uses these services scales up to handle more users. Services in a web services‐based
SOA tend to be coarse‐grained, document‐oriented, and asynchronous. As mentioned earlier, coarse‐grained
services offer a set of related business functions rather than a single function. For example, a coarse‐grained service
might handle the processing of a complete purchase order. By comparison, a fine‐grained service might handle only
one operation in the purchase order process. Again, as mentioned earlier, a document‐oriented service accepts a
document as input, as opposed to something more granular like a numeric value or Java object. An example of a
document‐oriented service might be a travel agency service that accepts as input a document that contains travel
information for a specific trip request. An asynchronous service performs its processing without forcing the client to
wait for the processing to finish. A synchronous service forces the client to wait. The relatively limited interaction
required for a client to communicate with a coarse‐grained, asynchronous service, especially a service that handles a
document such as a purchase order, allows applications that use these services to scale without putting a heavy
communication load on the network.
Flexibility: Loosely‐coupled services are typically more flexible than more tightly‐coupled applications. In a tightly‐
coupled architecture, the different components of an application are tightly bound to each other, sharing semantics,
libraries, and often sharing state. This makes it difficult to evolve the application to keep up with changing business
requirements. The loosely‐coupled, document‐based, asynchronous nature of services in an SOA allows applications
to be flexible, and easy to evolve with changing requirements.
Cost Efficiency: Other approaches that integrate disparate business resources such as legacy systems, business
partner applications, and department‐specific solutions are expensive because they tend to tie these components
together in a customized way. Customized solutions are costly to build because they require extensive analysis,
development time, and effort. They're also costly to maintain and extend because they're typically tightly‐coupled,
so that changes in one component of the integrated solution require changes in other components. A standards‐
based approach such as a web services‐based SOA should result in less costly solutions because the integration of
clients and services doesn't require the in‐depth analysis and unique code of customized solutions. Also, because
services in an SOA are loosely‐coupled, applications that use these services should be less costly to maintain and
easier to extend than customized solutions. In addition, a lot of the Web‐based infrastructure for a web services‐
based SOA is already in place in many enterprises, further limiting the cost. Last, but not least, SOA is about reuse of
business functions exposed as coarse‐grained services. This is potentially the biggest cost saving of all.
Most people tend to think of SOAP as nothing more than a protocol for Remote Procedure Calls (RPC) over Hypertext
Transfer Protocol (HTTP). However, this is only one implementation of SOAP, which is defined as a lightweight
protocol for passing structured and typed data between two peers using XML. The specification doesn’t require the
use of HTTP or even a request/response type of conversation. Instead, SOAP can be used with any protocol that
Vidyalankar Institute of Technology Page 40
supports the transmission of XML data from a sender to a receiver. In fact, both Microsoft and IBM have
implemented SOAP messages over SMTP, which means SOAP messages can be routed through e‐mail servers.
The bottom line is SOAP is nothing more than a lightweight messaging protocol, which can be used to send messages
between peers. The main goals of SOAP are focused on providing a common way to package message data and
define encoding rules used to serialize and deserialize the data during transmission. Another goal was to provide a
model that can be used to implement RPC operations using SOAP. All these goals are considered “orthogonal” in the
specification, which means they are independent, but related. For example: A SOAP message should define encoding
rules, but these rules needn’t be the same rules defined in the specification.
titleMy Life and Times/title
A related standard, SOAP Messages With Attachments, specifies the format of a SOAP message that includes
attachments (such as, images). SOAP messages (with or without attachments) are independent of any operating
system or platform and can be transported using a variety of communication protocols, such as HTTP or SMTP. WS‐I
Basic Profile 1.0 references SOAP 1.1.
How does a client know what format to use in making a request to a service For that matter, how do the client and
the service know what the request means The answers to these questions are provided by information in an XML
Vidyalankar Institute of Technology Page 41
document, called a WSDL document, that contains a description of the web service's interface. A WSDL document
contains information specified in Web Service Description Language (WSDL), as defined in the WSDL specification.
WSDL defines an XML schema for describing a web service. To uncover the description for a Web service, a client
needs to find the service's WSDL document. One way, perhaps the most typical way, to do this is for the client to find
a pointer to the WSDL document in the web service's registration, which can be in a UDDI registry or an ebXML
registry/repository. A typical scenario is that a business registers its service. The registry entry includes a pointer to a
WSDL file that contains the WSDL document for the service. Another business searches the registry and finds the
service. A programmer uses the interface information in the WSDL document to construct the appropriate calls to
the service. A WSDL document describes a web service as a collection of abstract items called "ports" or "endpoints."
A WSDL document also defines the actions performed by a web service and the data transmitted to these actions in
an abstract way. Actions are represented by "operations," and data is represented by "messages." A collection of
related operations is known as a "port type." A port type constitutes the collection of actions offered by a web
service. What turns a WSDL description from abstract to concrete is a "binding." A binding specifies the network
protocol and message format specifications for a particular port type. A port is defined by associating a network
address with a binding. If a client locates a WSDL document and finds the binding and network address for each port,
it can call the service's operations according to the specified protocol and message format.
Here, for example, is a WSDL document for an online book search service. Notice in the example that the WSDL
document specifies one operation for the service: getBooks:
operation name="getBooks" ...
The input message for the operation, BookSearchInput contains a string value named isbn:
The output message for the operation, BookSearchOutput returns a string value named title:
Notice too that the WSDL document specifies a SOAP protocol binding. The style attribute in the binding element
specifies that the receiver should interpret the payload in the SOAP message as an RPC method call:
Alternatively, the style attribute in a binding element could specify "document". In that case, a complete document
(typically an XML document) would be exchanged in the call. In fact, the document style of binding is more typical of
services in an SOA. One of the advantages of passing an XML document to a service instead of an RPC‐style message
is that it tends to be easier to validate and to apply business rules. The WS‐I Basic Profile 1.0 specifies the constructs
in WSDL 1.1 that be can be used to ensure interoperability. It also clarifies some of the construct descriptions in the
WSDL 1.1 specification.
UDDI and ebXML
As mentioned earlier, an SOA can also include a service that provides a directory or registry of services. But how can
a service be described in a registry so that a program looking for that service can easily find it and understand what it
Vidyalankar Institute of Technology Page 42
does The Universal Description, Discovery, and Integration (UDDI) specifications define how to publish and discover
information about services in a UDDI‐conforming registry. More specifically, the specifications define a UDDI schema
and a UDDI API. The UDDI schema identifies the types of XML data structures that comprise an entry in the registry
for a service. Think of a UDDI registry as a "Yellow Pages" for web services. Like the Yellow Pages directory for phone
numbers, a UDDI registry provides information about a service such as the name of the service, a brief description of
what it does, an address where the service can be accessed, and a description of the interface for accessing the
service. The accompanying figure illustrates the schema. It shows the five types of XML data structures that comprise
Here is an example that shows part of a complete BusinessEntity structure for a hypothetical company named
BooksToGo. The company provides various web services, including an online book ordering service.
The source for all professional books
The API describes the SOAP messages that are used to publish an entry in a registry, or discover an entry in a
registry. Here, for example, is a message that searches for all business entities in a registry whose name begins with
the characters "Books":
findbusiness generic="2.0" xmlns=um:uddiorg:apiv2"
UDDI 2.0 is part of the WS‐I Basic Profile 1.1
WS‐Security is a standard released by OASIS in March 2004. It describes security‐related enhancements to SOAP
messaging that provide for message integrity and confidentiality. Integrity means that a SOAP message is not
tampered with as it travels from a client to its final destination. Confidentiality means that a SOAP message is only
seen by intended recipients. WS‐Security uses security tokens to enable SOAP message security and integrity. A
security token is a collection of claims made by the sender of a SOAP message (such as the sender's identity). A
sender is authenticated by combining a security token with a digital signature – the signature is used as proof that
the sender is indeed associated with the security token. The standard also provides a general‐purpose mechanism
for associating security tokens with messages, and describes how to encode binary security tokens. WS‐Security is
quite flexible and can be used with a wide variety of security models and encryption technologies, such as Public‐key
infrastructure (PKI) and Kerberos, as well as the Secure Socket Layer (SSL)/ Transport Layer Security (TLS) protocol
that provides basic end‐to‐end security communication on the Internet.
(001) xml version="1.0" encoding="utf8"
(002) S:Envelope xmlns:S="http://www.w3.org/2001/12/soapenvelope"
Vidyalankar Institute of Technology Page 43
(004) m:path xmlns:m="http://schemas.xmlsoap.org/rp/"
(010) wsse:UsernameToken Id="MyID"
(017) ds:Reference URI="MsgBody"
(025) wsse:Reference URI="MyID"/
(031) S:Body Id="MsgBody"
(032) tru:StockSymbol xmlns:tru="http://fabrikam123.com/payloads"
The lines in the example that illustrate WS‐Security enhancements are:
• Line (009). This specifies a Security header that contains security information for an intended receiver.
This element continues until line (029).
• Lines (010) to (012). These lines specify a security token that is associated with the message. In this case, it
defines the username of the client using the UsernameToken element. The security token can also be used
to specify a password. However in this example it's assumed that the receiver knows the password ‐ in other
words, it is a shared secret.
• Lines (013) to (028) specify a digital signature. This signature ensures the integrity of the signed elements
(that they aren't modified). The signature follows the XML Signature specification. In this example, the
signature is based on a key generated from the user's password. Typically, stronger signing mechanisms
would be used.
• Lines (014) to (021) describe the digital signature. Line (015) specifies how to canonicalize (normalize) the
data that is being signed.
• Lines (017) to (020) select the elements that are signed. Specifically, line (017) indicates that the S:Body
element is signed. In this example, only the message body is signed. Typically, additional elements of the
message, such as parts of the routing header, should be included in the signature.
Vidyalankar Institute of Technology Page 44
• Line (022) specifies the signature value of the canonicalized form of the data that is being signed as defined
in the XML Signature specification.
• Lines (023) to (027) provide a hint as to where to find the security token associated with this signature.
Specifically, lines (024) to (025) indicate that the security token can be found at (pulled from) the specified
Vidyalankar Institute of Technology Page 45
Cluster of Servers
A computer cluster is a group of linked computers, working together closely thus in many respects forming a single
computer. The components of a cluster are commonly, but not always, connected to each other through fast local
Clusters are usually deployed to improve performance and/or availability over that of a single computer, while
typically being much more cost‐effective than single computers of comparable speed or availability.
• High‐availability (HA) clusters
o High‐availability clusters (also known as Failover
Clusters) are implemented primarily for the
purpose of improving the availability of services
that the cluster provides.
o They operate by having redundant nodes, which
are then used to provide service when system
o The most common size for an HA cluster is two
nodes, which is the minimum requirement to
o HA cluster implementations attempt to use
redundancy of cluster components to eliminate
single points of failure.
o There are commercial implementations of High‐
Availability clusters for many operating systems.
o One such implementation is the Gridlock platform
o The Linux‐HA project is one commonly used free software HA package for the Linux operating
o The LanderCluster from Lander Software can run on Windows, Linux, and UNIX platforms.
• Load‐balancing clusters
o Load‐balancing is when multiple computers are linked together to share computational workload or
function as a single virtual computer.
o Logically, from the user side, they are multiple machines, but function as a single virtual machine.
o Requests initiated from the user are managed by, and distributed among, all the standalone
computers to form a cluster.
o This results in balanced computational work among different machines, improving the performance
of the cluster systems.
• Compute Clusters
o Often clusters are used primarily for computational purposes, rather than handling IO‐oriented
operations such as web service or databases.
o For instance, a cluster might support computational simulations of weather or vehicle crashes.
o The primary distinction within compute clusters is how tightly‐coupled the individual nodes are.
o For instance, a single compute job may require frequent communication among nodes ‐ this implies
that the cluster shares a dedicated network, is densely located, and probably has homogenous
Vidyalankar Institute of Technology Page 46
o This cluster design is usually referred to as Beowulf Cluster.
o The other extreme is where a compute job uses one or few nodes, and needs little or no inter‐node
o This latter category is sometimes called "Grid" computing.
o Tightly‐coupled compute clusters are designed for work that might traditionally have been called
o Middleware such as MPI (Message Passing Interface) or PVM (Parallel Virtual Machine) permits
compute clustering programs to be portable to a wide variety of clusters.
• Grid Computing
o Grids are usually computer clusters, but more focused on throughput like a computing utility rather
than running fewer, tightly‐coupled jobs.
o Often, grids will incorporate heterogeneous collections of computers, possibly distributed
geographically, sometimes administered by unrelated organizations.
o Grid computing is optimized for workloads which consist of many independent jobs or packets of
work, which do not have to share data between the jobs during the computation process.
o Grids serve to manage the allocation of jobs to computers which will perform the work
independently of the rest of the grid cluster.
o Resources such as storage may be shared by all the nodes, but intermediate results of one job do not
affect other jobs in progress on other nodes of the grid.
o An example of a very large grid is the Foldinghome project.
o It is analyzing data that is used by researchers to find cures for diseases such as Alzheimer's and
o Another large project is the SETIhome project, which may be the largest distributed grid in
o It uses approximately three million home computers all over the world to analyze data from the
Arecibo Observatory radiotelescope, searching for evidence of extraterrestrial intelligence.
o In both of these cases, there is no inter‐node communication or shared storage.
o Individual nodes connect to a main, central location to retrieve a small processing job.
o They then perform the computation and return the result to the central server.
o In the case of the home projects, the software is generally run when the computer is otherwise
A cluster of servers allows servers to work together as computer cluster, to provide failover and increased
availability of applications, or parallel calculating power in case of high‐performance computing (HPC) clusters (as
A server cluster is a group of independent servers working together as a single system to provide high availability of
services for clients. When a failure occurs on one computer in a cluster, resources are redirected and the workload is
redistributed to another computer in the cluster.
You can use server clusters to ensure that users have constant access to important server‐based resources. Server
clusters are designed for applications that have long‐running in‐memory state or frequently updated data. Typical
uses for server clusters include file servers, print servers, database servers, and messaging servers.
A cluster consists of two or more computers working together to provide a higher level of availability, reliability, and
scalability than can be obtained by using a single computer. Microsoft cluster technologies guard against three
specific types of failure:
• Application and service failures, which affect application software and essential services.
Vidyalankar Institute of Technology Page 47
• System and hardware failures, which affect hardware components such as CPUs, drives, memory, network
adapters, and power supplies.
• Site failures in multisite organizations, which can be caused by natural disasters, power outages, or
The ability to handle failure allows server clusters to meet requirements for high availability, which is the ability to
provide users with access to a service for a high percentage of time while reducing unscheduled outages.
In a server cluster, each server owns and manages its local devices and has a copy of the operating system and the
applications or services that the cluster is managing. Devices common to the cluster, such as disks in common disk
arrays and the connection media for accessing those disks, are owned and managed by only one server at a time. For
most server clusters, the application data is stored on disks in one of the common disk arrays, and this data is
accessible only to the server that currently owns the corresponding application or service.
Server clusters are designed so that the servers in the cluster work together to protect data, keep applications and
services running after failure on one of the servers, and maintain consistency of the cluster configuration over time.
Types of Server Clusters
Single Quorum Device Cluster
The most widely used cluster type is the single
quorum device cluster, also called the standard
quorum cluster. In this type of cluster there are
multiple nodes with one or more cluster disk
arrays, also called the cluster storage, and a
connection device, that is, a bus. Each disk in the
array is owned and managed by only one server
at a time. The disk array also contains the
quorum resource. The following figure illustrates
a single quorum device cluster with one cluster
disk array. Because single quorum device clusters are the most widely used cluster, this Technical Reference focuses
on this type of cluster.
Majority Node Set Cluster
Windows Server 2003 supports another type of cluster, the
majority node set cluster. In a majority node set cluster, each
node maintains its own copy of the cluster configuration data.
The quorum resource keeps configuration data consistent
across the nodes. For this reason, majority node set clusters can
be used for geographically dispersed clusters. Another
advantage of majority node set clusters is that a quorum disk
can be taken offline for maintenance and the cluster as a whole
will continue to operate.
The major difference between majority node set clusters and single quorum device clusters is that single quorum
device clusters can operate with just one node, but majority node set clusters need to have a majority of the cluster
nodes available for the server cluster to operate. The following figure illustrates a majority node set cluster. For the
cluster in the figure to continue to operate, two of the three cluster nodes (a majority) must be available.
Vidyalankar Institute of Technology Page 48
Local Quorum Cluster
A local quorum cluster, also called a single node cluster, has a single node and is often used for testing. The following
figure illustrates a local quorum cluster.
Server virtualization is the masking of server resources, including the number and identity of individual physical
servers, processors, and operating systems, from server users. The server administrator uses a software application
to divide one physical server into multiple isolated virtual environments. The virtual environments are sometimes
called virtual private servers, but they are also known as guests, instances, containers or emulations.
Virtualization was invented more than thirty years ago to allow large expensive mainframes to be easily shared
among different application environments. As hardware prices went down, the need for virtualization faded away.
More recently, virtualization at all levels (system, storage, and network) became important again as a way to
improve system security, reliability and availability, reduce costs, and provide greater flexibility.
Approaches to Virtualization
There are three popular approaches to server virtualization:
1. The virtual machine model,
2. The paravirtual machine model,
3. Virtualization at the operating system (OS) layer.
Virtual machines are based on the host/guest paradigm. Each guest runs on a virtual imitation of the hardware
layer. This approach allows the guest operating system to run without modifications. It also allows the
administrator to create guests that use different operating systems. The guest has no knowledge of the host's
operating system because it is not aware that it's not running on real hardware.
It does, however, require real computing resources from the host ‐‐ so it uses a hypervisor to coordinate instructions
to the CPU. The hypervisor is called a virtual machine monitor (VMM). It validates all the guest‐issued CPU
instructions and manages any executed code that requires addition privileges. VMware and Microsoft Virtual Server
both use the virtual machine model.
Vidyalankar Institute of Technology Page 49
The paravirtual machine (PVM) model is also based on the host/guest paradigm ‐‐ and it uses a virtual machine
monitor too. In the paravirtual machine model, however, The VMM actually modifies the guest operating system's
code. This modification is called porting. Porting supports the VMM so it can utilize privileged systems calls
sparingly. Like virtual machines, paravirtual machines are capable of running multiple operating systems.
Virtualization OS Level
Virtualization at the OS level works a little differently. It isn't based on the host/guest paradigm. In the OS level
model, the host runs a single OS kernel as its core and exports operating system functionality to each of the guests.
Guests must use the same operating system as the host, although different distributions of the same system are
This distributed architecture eliminates system calls between layers, which reduces CPU usage overhead. It also
requires that each partition remain strictly isolated from its neighbors so that a failure or security breach in one
partition isn't able to affect any of the other partitions.
In this model, common binaries and libraries on the same physical machine can be shared, allowing an OS level
virtual server to host thousands of guests at the same time.
Guest Operating System Virtualization
In this scenario the physical host computer system runs a standard unmodified operating system such as Windows,
Linux, Unix or MacOS X. Running on this operating system is a virtualization application which executes in much the
same way as any other application such as a word processor or spreadsheet would run on the system. It is within this
virtualization application that one or more virtual machines are created to run the guest operating systems on the
The virtualization application is responsible for
starting, stopping and managing each virtual
machine and essentially controlling access to
physical hardware resources on behalf of the
individual virtual machines. The virtualization
application also engages in a process known as
binary rewriting which involves scanning the
instruction stream of the executing guest system
and replacing any privileged instructions with safe
This has the effect of making the guest system think
it is running directly on the system hardware, rather
than in a virtual machine within an application.
As outlined in the above diagram, the guest operating systems operate in virtual machines within the virtualization
application which, in turn, runs on top of the host operating system in the same way as any other application.
Clearly, the multiple layers of abstraction between the guest operating systems and the underlying host hardware
are not conducive to high levels of virtual machine performance.
This technique does, however, have the advantage that no changes are necessary to either host or guest operating
systems and no special CPU hardware virtualization support is required.
Vidyalankar Institute of Technology Page 50
Shared Kernel Virtualization
Shared kernel virtualization (also known as system level or operating system virtualization) takes advantage of the
architectural design of Linux and UNIX based operating systems. In order to understand how shared kernel
virtualization works it helps to first understand the two
main components of Linux or UNIX operating systems.
At the core of the operating system is the kernel. The
kernel, in simple terms, handles all the interactions
between the operating system and the physical
hardware. The second key component is the root file
system which contains all the libraries, files and
utilities necessary for the operating system to
function. Under shared kernel virtualization the virtual
guest systems each have their own root file system but
share the kernel of the host operating system. This
structure is illustrated in the following architectural
This type of virtualization is made possible by the ability of the kernel to dynamically change the current root file
system (a concept known as chroot) to a different root file system without having to reboot the entire system.
Essentially, shared kernel virtualization is an extension of this capability.
Perhaps the biggest single drawback of this form of virtualization is the fact that the guest operating systems must
be compatible with the version of the kernel which is being shared. It is not, for example, possible to run Microsoft
Windows as a guest on a Linux system using the shared kernel approach. Nor is it possible for a Linux guest system
designed for the 2.6 version of the kernel to share a 2.4 version kernel.
Kernel Level Virtualization
Under kernel level virtualization the host operating
system runs on a specially modified kernel which
contains extensions designed to manage and control
multiple virtual machines each containing a guest
operating system. Unlike shared kernel virtualization
each guest runs its own kernel, although similar
restrictions apply in that the guest operating systems
must have been compiled for the same hardware as the
kernel in which they are running. Examples of kernel
level virtualization technologies include User Mode
Linux (UML) and Kernel‐based Virtual Machine (KVM).
The following diagram provides an overview of the
kernel level virtualization architecture:
Vidyalankar Institute of Technology Page 51
Clo oud Computing
Clou ud computingg is Internet‐ ‐based comp puting, wherreby shared rresources, so oftware, and d information n are provide ed to
com mputers and o other device es on deman nd, like the e electricity grid. Cloud com mputing is a paradigm sh hift followingg the
shiftt from mainfframe to clie ent–server in n the early 1980s. Detailss are abstraccted from th he users, who o no longer have
need d for expertise in, or con ntrol over, the technologyy infrastructure "in the ccloud" that su upports them m.
Clou ud computin ng describess a new sup pplement, co onsumption,, and delive ery model fo or IT service es based on n the
Internet, and it ttypically invo olves over‐th he‐Internet p provision of dynamically scalable and d often virtualized resou urces.
It is a byproduct and consequence of th he ease‐of‐aaccess to rem mote compu uting sites prrovided by the Internet. This
frequently takess the form of web‐based d tools or applications th hat users can n access and d use through h a web browser
as iff it were a program installed locally o on their own computer.
NIST T provides aa somewhat more objecttive and specific definitio on here. The e term "cloud" is used ass a metapho or for
the Internet, bassed on the cloud drawingg used in the e past to represent the te elephone ne etwork and laater to depicct the
Internet in comp puter network diagrams as an abstraction of the underlying infrastructure e it represen nts.
Typiical cloud computing pro oviders delive er common b business app plications online that are e accessed fro om another Web
servvice or softw ware like a W Web browser,, while the ssoftware and d data are sttored on servvers. A key e element of ccloud
com mputing is cusstomization and the creaation of a use er‐defined exxperience.
Mosst cloud computing inffrastructuress consist off
servvices deliverred through h common centers and d
builtt on servers.
Clou uds often ap ppear as single points o of access forr
conssumers' com mputing need ds. Commerccial offeringss
are generally exxpected to meet qualitty of service e
(QoSS) requirem ments of cu ustomers, and typicallyy
inclu ude service level agreements (SLAss).The majorr
clou ud service prroviders include Salesforce, Amazon n
and Google. So ome of the larger IT firrms that are e
activvely involved d in cloud co omputing arre Microsoft,
Hew wlett Packard d and IBM.
Clou ud computingg derives chaaracteristics from, but sh hould not be confused with:
1. Autonom mic computing — "comp puter systems capable of self‐manage ement"
2. Client–server mode el – client– –server com mputing reffers broadlyy to any d distributed application that
distinguishes betwee en service prroviders (serrvers) and se ervice requessters (clients).
3. Grid com mputing — ""a form of d distributed computing an nd parallel ccomputing, w whereby a 'ssuper and virtual
compute er' is compo osed of a clu uster of netw worked, loosely coupled computers acting in con ncert to perform
very largge tasks"
4. Mainframe compute er — powerrful compute ers used maainly by largge organizatiions for critical applicattions,
typicallyy bulk data‐p processing su uch as censuss, industry and consume er statistics, e enterprise re esource planning,
and financial transacction processsing.
5. Utility co omputing — — the "packagging of comp puting resources, such ass computatio on and storage, as a mettered
service ssimilar to a traditional pu ublic utility, ssuch as electtricity";
6. Peer‐to‐ ‐peer – a disstributed arcchitecture wiithout the ne eed for centtral coordinaation, with participants b being
at the ssame time b both supplierrs and consu umers of resources (in contrast to the traditional client–se erver
Vidy yalankar Ins stitute of Technology Pag ge 52
In ge eneral, cloud d computingg customers do not own the physical infrastructu ure, instead avoiding cap pital expenditure
by renting usage e from a third‐party pro ovider. They consume resources as aa service and d pay only fo or resources that
theyy use. Manyy cloud‐com mputing offe erings emplo oy the utilitty computin ng model, w which is analogous to how
trad ditional utilityy services (su uch as electrricity) are con nsumed, whe ereas otherss bill on a sub bscription baasis.
Sharring "perishaable and inttangible" co omputing po ower among multiple te enants can iimprove utillization rates, as
servvers are not unnecessarilly left idle (w which can reduce costs ssignificantly w while increassing the speed of applicaation
A sid de‐effect of tthis approacch is that ove erall compute er usage rise es dramatically, as custom mers do not have to engiineer
for p peak load lim mits. In addittion, "increassed high‐spe eed bandwid dth" makes itt possible to receive the same. The ccloud
is be ecoming incrreasingly asssociated with h small and medium entterprises (SM MEs) as in maany cases they cannot ju ustify
or affford the largge capital exxpenditure of traditional IT.
SME Es also typicaally have lesss existing infrrastructure, less bureauccracy, more fflexibility, an nd smaller caapital budgetts for
purcchasing in‐ho ouse technollogy. Similarly, SMEs in e emerging maarkets are typ pically unburdened by esstablished le egacy
infraastructures, thus reducin ng the compllexity of deploying cloud solutions.
Clou ud architectu ure, the systtems archite ecture of the e software syystems involved in the d delivery of ccloud compu uting,
typiccally involve es multiple cloud components com mmunicatingg with each h other ove er applicatio on programming
interfaces, usually web serviices.
This resembles the Unix ph hilosophy off having
multtiple programs each do oing one thing well
and working toggether over u universal inte erfaces.
Com mplexity is controlled and the re esulting
systems are m more managgeable than n their
mon nolithic coun nterparts.
The two most significant co omponents o of cloud
com mputing archiitecture are known as th he front
end and the bacck end. The ffront end is tthe part
seen n by the client, i.e. the ccomputer usser. This
inclu udes the client’s networkk (or computter) and
the applicationss used to acccess the clou ud via a
userr interface su uch as a web b browser.
The back end of the cloud computing aarchitecture is the ‘cloud’ itself, com mprising various computters, serverss and
dataa storage devvices.
• Agility improves wiith users' ab bility to rapiidly and ine expensively rre‐provision technologiccal infrastruccture
• Cost is cclaimed to be greatly red duced and ccapital expen nditure is con nverted to o operational e expenditure. This
ostensib bly lowers baarriers to enttry, as infrasstructure is typically provvided by a th hird‐party an nd does not need
to be pu urchased for one‐time orr infrequent intensive co omputing tassks. Pricing o on a utility co omputing baasis is
fine‐graiined with usage‐based options and fewer IT skillss are require ed for implem mentation (in n‐house).
Vidy yalankar Ins stitute of Technology Pag ge 53
• Device and location independence enable users to access systems using a web browser regardless of their
location or what device they are using (e.g., PC, mobile). As infrastructure is off‐site (typically provided by a
third‐party) and accessed via the Internet, users can connect from anywhere.
• Multi‐tenancy enables sharing of resources and costs across a large pool of users thus allowing for:
o Centralization of infrastructure in locations with lower costs (such as real estate, electricity, etc.)
o Peak‐load capacity increases (users need not engineer for highest possible load‐levels)
o Utilization and efficiency improvements for systems that are often only 10–20 utilized.
• Reliability is improved if multiple redundant sites are used, which makes well designed cloud computing
suitable for business continuity and disaster recovery.38 Nonetheless, many major cloud computing
services have suffered outages, and IT and business managers can at times do little when they are affected.
• Scalability via dynamic ("on‐demand") provisioning of resources on a fine‐grained, self‐service basis near
real‐time, without users having to engineer for peak loads. Performance is monitored, and consistent and
loosely coupled architectures are constructed using web services as the system interface. One of the most
important new methods for overcoming performance bottlenecks for a large class of applications is data
parallel programming on a distributed data grid.
• Security could improve due to centralization of data, increased security‐focused resources, etc., but
concerns can persist about loss of control over certain sensitive data, and the lack of security for stored
kernels. Security is often as good as or better than under traditional systems, in part because providers are
able to devote resources to solving security issues that many customers cannot afford. Providers typically log
accesses, but accessing the audit logs themselves can be difficult or impossible. Furthermore, the complexity
of security is greatly increased when data is distributed over a wider area and / or number of devices.
• Maintenance of cloud computing applications is easier, since they don't have to be installed on each user's
computer. They are easier to support and to improve since the changes reach the clients instantly.
• Metering means that cloud computing resources usage should be measurable and should be metered per
client and application on a daily, weekly, monthly, and yearly basis.
Cloud computing provides the means through which everything — from computing power to computing
infrastructure, applications, business processes to personal collaboration — can be delivered to you as a service
wherever and whenever you need.
Cloud computing is offered in different forms:
• Public clouds
o Public cloud or external cloud describes cloud computing in the traditional mainstream sense,
whereby resources are dynamically provisioned on a fine‐grained, self‐service basis over the
Internet, via web applications/web services, from an off‐site third‐party provider who bills on a fine‐
grained utility computing basis.
• Community Clouds
o A community cloud may be established where several organizations have similar requirements and
seek to share infrastructure so as to realize some of the benefits of cloud computing. With the costs
spread over fewer users than a public cloud (but more than a single tenant) this option is more
expensive but may offer a higher level of privacy, security and/or policy compliance.
• Private clouds
• Hybrid clouds
o A hybrid cloud environment consisting of multiple internal and/or external providers "will be typical
for most enterprises". By integrating multiple cloud services users may be able to ease the transition
to public cloud services.
Vidyalankar Institute of Technology Page 54
Cloud Computing Layers
Infrastructure As A Service (IAAS)
IaaS is the delivery of computer hardware (servers, networking technology, storage, and data centre space) as a
service. You also can expect it to include the delivery of operating systems and virtualization technology to manage
the resources. The IaaS customers rent computing resources instead of buying and installing them in their own data
centre. Consumers control and manage the systems in terms of the operating systems, applications, storage, and
network connectivity, but do not themselves control the cloud infrastructure.
• Much of the IaaS market will likely follow in the path of the ISP market. The ISP model has been proven and
some large ISPs (GoDaddy at www.godaddy. com and inMotion hosting at www.inmotionhosting.com) run
millions of Web sites.
• The ISP service is typically paid for based on the amount of resources used over time. This can include
dynamic scaling so that if more resources are required than expected, they will be provided immediately (up
to a given limit).
• The arrangement involves an agreed‐upon service level — normally 99.9 percent availability or better, with
limits set on CPU usage, memory, disk space, and Internet bandwidth. No one will object if you want to rent
a server or a virtual server from an ISP and you run a data mart (instead of running a Web site).
• Nothing in the customer agreement stops you from using the resources in that way. It wouldn’t make much
sense, however, because you probably wouldn’t get the service level agreement you wanted or the support
you needed. IaaS takes the ISP model to a new level.
Platform As A Service (PAAS)
PaaS has many definitions, we’d like you to think about it as a computing platform that includes a set of
development, middleware, and deployment capabilities. A key vendor characteristic is creating and encouraging a
deep ecosystem of partners who all commit to this environment for the future.
When organizations are looking for capacity on demand, they often look to Infrastructure as a Service (IaaS).
However, when an organization is looking for a deeper set of capabilities, they look at Platform as a Service (PaaS).
• PaaS has to leverage the Internet.
• PaaS must offer some type of development language so professional developers (and in some cases users)
can add value.
Vidyalankar Institute of Technology Page 55
• These environments need a way to monitor and measure resource use and to track overall performance of
the vendor’s platform.
• Almost all PaaS platforms are based on a multi‐tenancy architecture (which lets multiple clients run their
copy separately from each other through virtualization) so that each customer’s code or data is isolated from
• A PaaS environment needs to support the development lifecycle and the team development process,
• A PaaS platform needs to include services interfaces such as SOAP (Simple Object Access Protocol) and XML
(eXtensible Markup Language), among others.
• A PaaS platform must be able to deploy, manage, test, and maintain the developed applications.
• A PaaS platform must support well‐defined and well‐documented interfaces so elements and components
can be used in the following:
o Composite applications are created by combining services to create an enterprise application based
on orchestration of business logic and rules.
o Portals, which are an organized environment that organizes application components for the
o Mashups, which let end users easily bring together two or more business services that can
communicate and exchange data.
Software As A Service (SAAS)
Mainframe systems were simply too expensive for most companies to buy their own systems. A couple of decades
later, minicomputers, servers, and personal computers changed the dynamics of the market. Economically, it was
feasible for any Tom, Dick, and Harriet to own their own systems and the software. Not all software moved to an
internal model however. (Software such as ADP’s payroll system, for example, remained Software as a Service.)
Two key events converged to create the model that we now call Software as a Service (SaaS):
• First, the Internet became a commercial platform.
• Second, software costs and complexities became so difficult that running, upgrading, and managing software
become too complex for many companies to manage.
o This was especially true for small‐ and medium sized companies that didn’t want the expenses of
managing all the components. These companies were the first to embrace this new generation of
Characterizing Software as a Service:
What characteristics have to be in place for a SaaS to be commercially viable Here’s what we think is necessary:
• The SaaS application needs to be generalized enough so that lots of customers will be interested in the
service. Here are some examples of these types of applications: accounting, collaboration, project
management, testing, analytics, content management, Internet marketing, risk management and of course,
CRM. What doesn’t work as SaaS A specialized one of‐ a‐kind application with a small number of potential
• SaaS applications need sophisticated navigation and ease of use. If an SaaS application isn’t easy to use,
customers will simply stop subscribing. Most SaaS vendors offer prospective customers a free trial for a
month or so. If the customer doesn’t start using the application during that first month, it’s likely that the
customer won’t sign a contract. This is really important because it has been reported that less than 20
percent of users remain customers after the first month or so.
• The SaaS application needs be modular and service oriented. Without this modular approach, it will be hard
to change and difficult to have third‐party independent companies join the ecosystem.
• An SaaS application needs to include measuring and monitoring so customers can be charged actual usage.
Vidyalankar Institute of Technology Page 56
• An SaaS application must have a built‐in billing service.
• SaaS applications need published interfaces and an ecosystem of partners who can expand the company’s
customer base and market reach.
• SaaS applications have to ensure that each customer’s data and specialized configurations are separate
and secure from other customers’ data and configurations.
• SaaS applications need to provide sophisticated business process configurators for customers. Each
customer can change the process within the standardized SaaS application. For example, a company might
want to add a process so a manager has to approve the price being offered to a new customer. A built‐in
configuration tool enables this to be done on an ad hoc basis without programming.
• SaaS applications need to constantly provide fast releases of new features and new capabilities. This must
be done without impacting the customer’s ability to continue business as usual.
• SaaS applications have to protect the integrity of customer data.
• That includes providing techniques for allowing data to migrate either to a private database inside the
firewall or to a third‐party storage capability.
Vidyalankar Institute of Technology Page 57
Server Consolidation using Cloud
Server consolidation is an approach to the efficient usage of computer server resources in order to reduce the total
number of servers or server locations that an organization requires. The practice developed in response to the
problem of server sprawl, a situation in which multiple, under‐utilized servers take up more space and consume
more resources than can be justified by their workload.
• Servers in many companies typically run at 15‐20 of their capacity,
o which may not be a sustainable ratio in the current economic environment.
• Businesses are increasingly turning to server consolidation as one means of cutting unnecessary costs and
maximizing return on investment (ROI) in the data center.
o Of 518 respondents in a Gartner Group research study, six percent had conducted a server
o 61 were currently conducting one, and
o 28 were planning to do so in the immediate future.
Although consolidation can substantially increase the efficient use of server resources, it may also result in complex
configurations of data, applications, and servers that can be confusing for the average user to contend with. To
alleviate this problem, server virtualization may be used to mask the details of server resources from users while
optimizing resource sharing. Another approach to server consolidation is the use of blade servers to maximize the
efficient use of space.
There are four things to consider when looking at server consolidation: hardware, redundancy, operating system,
and maximizing efficiency.
Cloud computing can be used for Server Consolidation by moving applications such as Exchange Mail, Data Storage
etc. to a public cloud.
Note: Relate all the cloud layers and deployment models to server consolidation and explain the applications that
can be housed remotely on a cloud to minimize the number, redundancy and underutilization of servers.
For many companies, the concept of server consolidation connotes sweeping projects that replace numerous smaller
servers with powerful, expensive workhorses. Many organizations have pursued this approach and realized
significant cost savings and improved efficiency. Achieving the benefits of server consolidation, however, does not
necessarily require a large‐scale effort that is both extensive and expensive in terms of a lengthy planning cycle and
up‐front investment. Many organizations consolidate as part of the natural refresh cycle of their technology
infrastructures. These efforts are more iterative processes, rather than one‐time projects, that focus on reducing the
number of datacenters as the technology infrastructure is updated and augmented. The results are still significant.
By consolidating, organizations are able to boost efficiency and improve their service capabilities while maintaining
an infrastructure that is both robust, highly available, and adaptive to changing business requirements. Dell is
introducing an expanded program aimed at consolidating Microsoft‐based environments to help customers reduce
their total costs and gain greater control over their IT infrastructure. The company hopes to use its new Server
Consolidation ROI Analyst Tool, latest‐generation enterprise servers, systems management software, and services to
help organizations realize the benefits of consolidation as they renew their technology infrastructures.
Drivers Benefits of Server Consolidation
A good deal of the impetus for consolidation stems from the proliferation of distributed systems that occurred
during the 1980s and 1990s. As the price/performance and functionality of small and midrange systems improved
dramatically, many organizations moved toward a model of deploying distributed servers to support departmental
Vidyalankar Institute of Technology Page 58
While such systems allowed organizations relatively cheap and easy access to powerful departmental applications,
they eventually led to a quagmire of systems that were incompatible, unable to share information, and difficult or
impossible to manage consistently. Organizations also faced the challenge of meeting increased levels of availability
and security. At the same time, many companies looked to standardize their environments on a consolidated set of
products from a more limited number of vendors in order to realize not only cost savings but also procurement,
service, and management efficiencies.
At the same time, many IT organizations became more service oriented. Their focus shifted from managing
processes and technology to providing information access to a broad set of user constituencies. Users began to
expect universal access to applications and information.
Companies found that the duplication of servers, applications, and databases not only was expensive to maintain but
also kept them from utilizing their resources and information to the maximum extent.
The response has been a trend toward recentralization and consolidation of IT resources in the datacenter.
Consolidation allows companies to improve overall business processing through three primary IT objectives: a higher
and more consistent level of service, greater efficiency and control over operations, and the flexibility to respond to
constantly changing business requirements (see Figure 1).
ACHIEVING HIGHER LEVELS OF SERVICE
Providing high levels of service is on every IT manager's mind as the user base expands beyond internal users to
customers, suppliers, students, patients, other government agencies, and business partners.
Everyone expects high levels of application and information availability and consistent response times. Consolidation
provides a consistent management framework, which can lead to a more predictable and consistent level of service.
EFFICIENCY AND CONTROL
The key to gaining efficiency is in better utilizing the available IT resources. Consolidating the server environment can
lead to a more disciplined approach to management. As much as 55 of IT costs are associated with personnel, and,
in a consolidated environment, the productivity of administrative personnel increases greatly.
Vidyalankar Institute of Technology Page 59
Some organizations may realize the benefit in reduced operational costs. IDC studies have shown a 7:1 cost savings
in people management resources when processes and resources are consolidated. Others will benefit from the
ability to focus highly skilled resources on higher‐value tasks. Tighter management control will also enhance
availability, security, and the ability to audit the consumption of services and charge back appropriately.
For IT managers who consistently cite floor space and power consumption as key issues, the benefits of
consolidation are obvious. Many organizations are looking to reduce not only the number of servers but the number
of datacenters. Because the cost of network bandwidth has dropped dramatically, companies no longer need to
deploy servers as close as possible to the users they support. Therefore, many organizations can implement smaller,
regional datacenters, instead of a datacenter at each location.
As server architectures grow increasingly modular, with the deployment of ultra‐dense solutions such as server
blades, the costs associated with power, space, cooling, and cabling will continue to drop.
GREATER ORGANIZATIONAL FLEXIBILITY
The greatest strategic value that companies gain from consolidation is an improved ability to efficiently adapt the
infrastructure to incorporate new technologies and respond to new business requirements. All types of
organizations — business, government, medical providers, and educational institutions — are driven by the need for
high levels of services and information availability. With a consistent framework for managing data, IT spends less
time moving data and transforming it into a usable form and, therefore, can respond more quickly to demands for
data availability. The infrastructure is then better able to adapt as the organization moves forward. Once the IT
center has completed the consolidation process and achieved the desired environment, users can expect more rapid
deployment of new applications and features, leading to greater flexibility to respond to changing demands.
TYPES OF CONSOLIDATION
Consolidation of the server environment can occur at many different levels. Most organizations pursue different
levels of consolidation at different times, depending on their particular requirements. Consolidation also tends to
occur over time, as an iterative process, moving through the different types of consolidation.
The various types of consolidation should be viewed as steps along a continuum. The degrees of consolidation can
be described in four general categories (see Figure 2):
→Logical consolidation reduces the number of points of control in the environment to a single administrative
stream and reduces the number of consoles. The servers remain dispersed, while local operations are reduced or
eliminated and management functions such as backup, restore, recovery, maintenance, and user support are
performed remotely. A major benefit of logical consolidation is a reduction in operational headcount, or more
efficient use of the skills already on hand. Using fewer people and a consistent set of products and processes can
help reduce both the dilution of skills across the organization and the opportunity for errors. This approach reduces
the cost of maintaining the environment and improves service to users.
→Physical consolidation involves consolidating components of the IT environment in one physical location. This
colocation leads to greater efficiency by eliminating the replication of skill sets across different locations. When
systems are in a central location, networking becomes much easier and more efficient, power costs are reduced,
backup can be performed more efficiently, and security can be increased. There are also subsequent savings in floor
space costs. Consolidating locations also provides opportunity for configuring systems for higher availability.
→Workload consolidation reduces the number and variety of components in the environment. It involves not only
servers but also other physical elements such as tapes, disks, network devices and connections, software, operating
systems, and peripherals. The number of processes and procedures are also reduced. With fewer hardware and
software standards to manage, IT departments can more easily move and change systems, applications, and
peripherals. Network performance is enhanced even further, as are security and availability. Common operating
Vidyalankar Institute of Technology Page 60
system environments allow more applications to share the same server, taking advantage of faster processors.
Workload consolidation can be organized by application type, by operating system type, or by line of business.
→Transparent consolidation (including storage) involves pulling together a number of IT centers across a campus or
network and implementing storage area networks to create a single set of resources. These environments will be
highly automated, and their growth will be driven by the availability of high‐speed, low‐cost data networks.
Vidyalankar Institute of Technology Page 61
Google Apps Engine
Google App Engine is a web application hosting service. By “web application,” we mean an application or service
accessed over the Web, usually with a web browser: storefronts with shopping carts, social networking sites,
multiplayer games, mobile applications, survey applications, project management, collaboration, publishing, and all
of the other things we’re discovering are good uses for the Web. App Engine can serve traditional website content
too, such as documents and images, but the environment is especially designed for real‐time dynamic applications.
In particular, Google App Engine is designed to host applications with many simultaneous users. When an application
can serve many simultaneous users without degrading performance, we say it scales. Applications written for App
Engine scale automatically. As more people use the application, App Engine allocates more resources for the
application and manages the use of those resources. The application itself does not need to know anything about
the resources it is using.
Unlike traditional web hosting or self‐managed servers, with Google App Engine, you only pay for the resources you
use. These resources are measured down to the gigabyte, with no monthly fees or up‐front charges. Billed resources
include CPU usage, storage per month, incoming and outgoing bandwidth, and several resources specific to App
To help you get started, every developer gets a certain amount of resources for free, enough for small applications
with low traffic. Google estimates that with the free resources, an app can accommodate about 5 million page views
App Engine can be described as three parts: the runtime environment, the datastore, and the scalable services. In
this chapter, we’ll look at each of these parts at a high level.
The Runtime Environment
An App Engine application responds to web requests. A web request begins when a client, typically a user’s web
browser, contacts the application with an HTTP request, such as to fetch a web page at a URL. When App Engine
receives the request, it identifies the application from the domain name of the address, either an .appspot.com
subdomain (provided for free with every app) or a subdomain of a custom domain name you have registered and set
up with Google Apps. App Engine selects a server from many possible servers to handle the request, making its
selection based on which server is most likely to provide a fast response. It then calls the application with the
content of the HTTP request, receives the response data from the application, and returns the response to the client.
From the application’s perspective, the runtime environment springs into existence when the request handler
begins, and disappears when it ends. App Engine provides at least two methods for storing data that persists
between requests (discussed later), but these mechanisms live outside of the runtime environment. By not retaining
state in the runtime environment between requests—or at least, by not expecting that state will be retained
between requests—App Engine can distribute traffic among as many servers as it needs to give every request the
same treatment, regardless of how much traffic it is handling at one time.
Application code cannot access the server on which it is running in the traditional sense. An application can read its
own files from the filesystem, but it cannot write to files, and it cannot read files that belong to other applications.
An application can see environment variables set by App Engine, but manipulations of these variables do not
necessarily persist between requests. An application cannot access the networking facilities of the server hardware,
though it can perform networking operations using services.
In short, each request lives in its own “sandbox.” This allows App Engine to handle a request with the server that
would, in its estimation, provide the fastest response. There is no way to guarantee that the same server hardware
will handle two requests, even if the requests come from the same client and arrive relatively quickly. Sandboxing
Vidyalankar Institute of Technology Page 62
also allows App Engine to run multiple applications on the same server without the behaviour of one application
In addition to limiting access to the operating system, the runtime environment also limits the amount of clock time,
CPU use, and memory a single request can take. App Engine keeps these limits flexible, and applies stricter limits to
applications that use up more resources to protect shared resources from “runaway” applications. A request has up
to 30 seconds to return a response to the client. While that may seem like a comfortably large amount for a web
app, App Engine is optimized for applications that respond in less than a second.
Also, if an application uses many CPU cycles, App Engine may slow it down so the app isn’t hogging the processor on
a machine serving multiple apps. A CPU‐intensive request handler may take more clock time to complete.
Most useful web applications need to store information during the handling of a request for retrieval during a later
request. A typical arrangement for a small website involves a single database server for the entire site, and one or
more web servers that connect to the database to store or retrieve data. Using a single central database server
makes it easy to have one canonical representation of the data, so multiple users accessing multiple web servers all
see the same and most recent information.
But a central server is difficult to scale once it reaches its capacity for simultaneous connections. By far the most
popular kind of data storage system for web applications in the past decade has been the relational database, with
tables of rows and columns arranged for space efficiency and concision, and with indexes and raw computing power
for performing queries, especially “join” queries that can treat multiple related records as a queryable unit.
Other kinds of data storage systems include hierarchical datastores (file systems, XML databases) and object
databases. Each kind of database has pros and cons, and which type is best suited for an application depends on the
nature of the application’s data and how it is accessed. And each kind of database has its own techniques for
growing past the first server.
Google App Engine’s database system most closely resembles an object database. It is not a join‐query relational
database, and if you come from the world of relational‐database‐backed web applications (as I did), this will
probably require changing the way you think about your application’s data. As with the runtime environment, the
design of the App Engine datastore is an abstraction that allows App Engine to handle the details of distributing and
scaling the application, so your code can focus on other things.
The datastore’s relationship with the runtime environment is that of a service: the application uses an API to access a
separate system that manages all of its own scaling needs separately from the runtime environment. Google App
Engine includes several other self‐scaling services useful for web applications.
The memory cache (or memcache) service is a short‐term key‐value storage service. Its main advantage over the
datastore is that it is fast, much faster than the datastore for simple storage and retrieval. The memcache stores
values in memory instead of on disk for faster access. It is distributed like the datastore, so every request sees the
same set of keys and values. However, it is not persistent like the datastore: if a server goes down, such as during a
power failure, memory is erased. It also has a more limited sense of atomicity and transactionality than the
datastore. As the name implies, the memcache service is best used as a cache for the results of frequently
performed queries or calculations.
The application checks for a cached value, and if the value isn’t there, it performs the query or calculation and stores
the value in the cache for future use. App Engine applications can access other web resources using the URL Fetch
Vidyalankar Institute of Technology Page 63
The service makes HTTP requests to other servers on the Internet, such as to retrieve pages or interact with web
services. Since remote servers can be slow to respond, the URL Fetch API supports fetching URLs in the background
while a request handler does other things, but in all cases the fetch must start and finish within the request handler’s
lifetime. The application can also set a deadline, after which the call is cancelled if the remote host hasn’t responded.
App Engine applications can send messages using the Mail service. Messages can be sent on behalf of the application
or on behalf of the user who made the request that is sending the email (if the message is from the user). Many web
applications use email to notify users, confirm user actions, and validate contact information.
An application can also receive email messages. If an app is configured to receive email, a message sent to the app’s
address is routed to the Mail service, which delivers the message to the app in the form of an HTTP request to a
App Engine applications can send and receive instant messages to and from chat services that support the XMPP
protocol, including Google Talk. An app sends an XMPP chat message by calling the XMPP service. As with incoming
email, when someone sends a message to the app’s address, the XMPP service delivers it to the app by calling a
Vidyalankar Institute of Technology Page 64
We live in the data age. It's not easy to measure the total volume of data stored electronically, but an IDC estimate
put the size of the "digital universe" at 0.18 zettabytes in 2006, and is forecasting a tenfold growth by 2011 to 1.8
zettabytes. A zettabyte is 10 bytes, or equivalently one thousand exabytes, one million petabytes, or one billion
terabytes. That's roughly the same order of magnitude as one disk drive for every person in the world. This flood of
data is coming from many sources. Consider the following:
• The New York Stock Exchange generates about one terabyte of new trade data per day.
• Facebook hosts approximately 10 billion photos, taking up one petabyte of storage.
• Ancestry.com, the genealogy site, stores around 2.5 petabytes of data.
• The Internet Archive stores around 2 petabytes of data, and is growing at a rate of 20 terabytes per month.
• The Large Hadron Collider near Geneva, Switzerland, will produce about 15 petabytes of data per year.
So there's a lot of data out there. But you are probably wondering how it affects you. Most of the data is locked up in
the largest web properties (like search engines), or scientific or financial institutions, isn't it Does the advent of "Big
Data," as it is being called, affect smaller organizations or individuals
Arguably it does. Take photos, for example. My wife's grandfather was an avid photographer, and took photographs
throughout his adult life. His entire corpus of medium format, slide, and 35mm film, when scanned in at high‐
resolution, occupies around 10 gigabytes. Compare this to the digital photos that my family took last year, which
take up about 5 gigabytes of space. My family is producing photographic data at 35 times the rate my wife's
grandfather's did, and the rate is increasing every year as it becomes easier to take more and more photos.
More generally, the digital streams that individuals are producing are growing apace. Microsoft Research's
MyLifeBits project gives a glimpse of archiving of personal information that may become commonplace in the near
future. MyLifeBits was an experiment where an individual's interactions—phone calls, emails, documents—were
captured electronically and stored for later access. The data gathered included a photo taken every minute, which
resulted in an overall data volume of one gigabyte a month. When storage costs come down enough to make it
feasible to store continuous audio and video, the data volume for a future MyLifeBits service will be many times
The trend is for every individual's data footprint to grow, but perhaps more importantly the amount of data
generated by machines will be even greater than that generated by people. Machine logs, RFID readers, sensor
networks, vehicle GPS traces, retail transactions—all of these contribute to the growing mountain of data.
The volume of data being made publicly available increases every year too. Organizations no longer have to merely
manage their own data: success in the future will be dictated to a large extent by their ability to extract value from
other organizations' data.
Initiatives such as Public Data Sets on Amazon Web Services, Infochimps.org, and theinfo.org exist to foster the
"information commons," where data can be freely (or in the case of AWS, for a modest price) shared for anyone to
download and analyze. Mashups between different information sources make for unexpected and hitherto
Take, T for example, the Astrometry.net project, which watches the Astrometry group on Flickr for new photos of the
night sky. It analyzes each image, and identifies which part of the sky it is from, and any interesting celestial bodies,
such as stars or galaxies. Although it's still a new and experimental service, it shows the kind of things that are
possible when data (in this case, tagged photographic images) is made available and used for something (image
analysis) that was not anticipated by the creator.
Vidyalankar Institute of Technology Page 65
It has been said that "More data usually beats better algorithms," which is to say that for some problems (such as
recommending movies or music based on past preferences), however fiendish your algorithms are, they can often be
beaten simply by having more data (and a less sophisticated algorithm). The good news is that Big Data is here. The
bad news is that we are struggling to store and analyze it.
Data Storage and Analysis
The problem is simple: while the storage capacities of hard drives have increased massively over the years, access
speeds—the rate at which data can be read from drives— have not kept up. One typical drive from 1990 could store
1370 MB of data and had a transfer speed of 4.4 MB/s,§ so you could read all the data from a full drive in around five
minutes. Almost 20 years later one terabyte drives are the norm, but the transfer speed is around 100 MB/s, so it
takes more than two and a half hours to read all the data off the disk.
This is a long time to read all data on a single drive—and writing is even slower. The obvious way to reduce the time
is to read from multiple disks at once. Imagine if we had 100 drives, each holding one hundredth of the data.
Working in parallel, we could read the data in under two minutes.
Only using one hundredth of a disk may seem wasteful. But we can store one hundred datasets, each of which is one
terabyte, and provide shared access to them. We can imagine that the users of such a system would be happy to
share access in return for shorter analysis times, and, statistically, that their analysis jobs would be likely to be
spread over time, so they wouldn't interfere with each other too much.
There's more to being able to read and write data in parallel to or from multiple disks, though.
The first problem to solve is hardware failure: as soon as you start using many pieces of hardware, the chance that
one will fail is fairly high. A common way of avoiding data loss is through replication: redundant copies of the data
are kept by the system so that in the event of failure, there is another copy available. This is how RAID works, for
instance, although Hadoop's filesystem, the Hadoop Distributed Filesystem (HDFS), takes a slightly different
approach, as you shall see later.
The second problem is that most analysis tasks need to be able to combine the data in some way; data read from
one disk may need to be combined with the data from any of the other 99 disks. Various distributed systems allow
data to be combined from multiple sources, but doing this correctly is notoriously challenging. MapReduce provides
a programming model that abstracts the problem from disk reads and writes, transforming it into a computation
over sets of keys and values. The important point for the present discussion is that there are two parts to the
computation, the map and the reduce, and it's the interface between the two where the "mixing" occurs. Like HDFS,
MapReduce has reliability built‐in.
This, in a nutshell, is what Hadoop provides: a reliable shared storage and analysis system. The storage is provided by
HDFS, and analysis by MapReduce. There are other parts to Hadoop, but these capabilities are its kernel.
Comparison with Other Systems
The approach taken by MapReduce may seem like a brute‐force approach. The premise is that the entire dataset—or
at least a good portion of it—is processed for each query. But this is its power. MapReduce is a batch query
processor, and the ability to run an ad hoc query against your whole dataset and get the results in a reasonable time
is transformative. It changes the way you think about data, and unlocks data that was previously archived on tape or
disk. It gives people the opportunity to innovate with data. Questions that took too long to get answered before can
now be answered, which in turn leads to new questions and new insights.
For example, Mailtrust, Rackspace's mail division, used Hadoop for processing email logs. One ad hoc query they
wrote was to find the geographic distribution of their users. In their words:
Vidyalankar Institute of Technology Page 66
This data was so useful that we've scheduled the MapReduce job to run monthly and we will be using this data to
help us decide which Rackspace data centers to place new mail servers in as we growJ
By bringing several hundred gigabytes of data together and having the tools to analyze it, the Rackspace engineers
were able to gain an understanding of the data that they otherwise would never have had, and, furthermore, they
were able to use what they had learned to improve the service for their customers.
Why can't we use databases with lots of disks to do large‐scale batch analysis Why is MapReduce needed
The answer to these questions comes from another trend in disk drives: seek time is improving more slowly than
transfer rate. Seeking is the process of moving the disk's head to a particular place on the disk to read or write data.
It characterizes the latency of a disk operation, whereas the transfer rate corresponds to a disk's bandwidth.
If the data access pattern is dominated by seeks, it will take longer to read or write large portions of the dataset than
streaming through it, which operates at the transfer rate. On the other hand, for updating a small proportion of
records in a database, a traditional B‐Tree (the data structure used in relational databases, which is limited by the
rate it can perform seeks) works well. For updating the majority of a database, a B‐Tree is less efficient than
MapReduce, which uses Sort/Merge to rebuild the database.
In many ways, MapReduce can be seen as a complement to an RDBMS. (The differences between the two systems
are shown in Table 1‐1.) MapReduce is a good fit for problems that need to analyze the whole dataset, in a batch
fashion, particularly for ad hoc analysis. An RDBMS is good for point queries or updates, where the dataset has been
indexed to deliver low‐latency retrieval and update times of a relatively small amount of data. MapReduce suits
applications where the data is written once, and read many times, whereas a relational database is good for datasets
that are continually updated.
Table 1‐1. RDBMS compared to MapReduce
Traditional RDBMS MapReduce
Data size Gigabytes Petabytes
Access Interactive and batch Batch
Updates Read and write many times Write once, read many times
Structure Static schema Dynamic schema
Integrity High Low
Scaling Nonlinear Linear
Another difference between MapReduce and an RDBMS is the amount of structure in the datasets that they operate
on. Structured data is data that is organized into entities that have a defined format, such as XML documents or
database tables that conform to a particular predefined schema. This is the realm of the RDBMS. Semi‐structured
data, on the other hand, is looser, and though there may be a schema, it is often ignored, so it may be used only as a
guide to the structure of the data: for example, a spreadsheet, in which the structure is the grid of cells, although the
cells themselves may hold any form of data. Unstructured data does not have any particular internal structure: for
example, plain text or image data. MapReduce works well on unstructured or semi‐structured data, since it is
Vidyalankar Institute of Technology Page 67
designed to interpret the data at processing time. In other words, the input keys and values for MapReduce are not
an intrinsic property of the data, but they are chosen by the person analyzing the data.
Relational data is often normalized to retain its integrity, and remove redundancy. Normalization poses problems for
MapReduce, since it makes reading a record a nonlocal operation, and one of the central assumptions that
MapReduce makes is that it is possible to perform (high‐speed) streaming reads and writes.
A web server log is a good example of a set of records that is not normalized (for example, the client hostnames are
specified in full each time, even though the same client may appear many times), and this is one reason that logfiles
of all kinds are particularly well‐suited to analysis with MapReduce.
MapReduce is a linearly scalable programming model. The programmer writes two functions—a map function and a
reduce function—each of which defines a mapping from one set of key‐value pairs to another. These functions are
oblivious to the size of the data or the cluster that they are operating on, so they can be used unchanged for a small
dataset and for a massive one. More importantly, if you double the size of the input data, a job will run twice as
slow. But if you also double the size of the cluster, a job will run as fast as the original one. This is not generally true
of SQL queries.
Over time, however, the differences between relational databases and MapReduce systems are likely to blur. Both
as relational databases start incorporating some of the ideas from MapReduce (such as Aster Data's and
Greenplum's databases), and, from the other direction, as higher‐level query languages built on MapReduce (such as
Pig and Hive) make MapReduce systems more approachable to traditional database programmers.
The High Performance Computing (HPC) and Grid Computing communities have been doing large‐scale data
processing for years, using such APIs as Message Passing Interface (MPI). Broadly, the approach in HPC is to
distribute the work across a cluster of machines, which access a shared filesystem, hosted by a SAN. This works well
for predominantly compute‐intensive jobs, but becomes a problem when nodes need to access larger data volumes
(hundreds of gigabytes, the point at which MapReduce really starts to shine), since the network bandwidth is the
bottleneck, and compute nodes become idle.
MapReduce tries to colocate the data with the compute node, so data access is fast since it is local. This feature,
known as data locality, is at the heart of MapReduce and is the reason for its good performance. Recognizing that
network bandwidth is the most precious resource in a data center environment (it is easy to saturate network links
by copying data around), MapReduce implementations go to great lengths to preserve it by explicitly modelling
network topology. Notice that this arrangement does not preclude high‐CPU analyses in MapReduce.
MPI gives great control to the programmer, but requires that he or she explicitly handle the mechanics of the data
flow, exposed via low‐level C routines and constructs, such as sockets, as well as the higher‐level algorithm for the
analysis. MapReduce operates only at the higher level: the programmer thinks in terms of functions of key and value
pairs, and the data flow is implicit.
Coordinating the processes in a large‐scale distributed computation is a challenge. The hardest aspect is gracefully
handling partial failure—when you don't know if a remote process has failed or not—and still making progress with
the overall computation. MapReduce spares the programmer from having to think about failure, since the
implementation detects failed map or reduce tasks and reschedules replacements on machines that are healthy.
MapReduce is able to do this since it is a shared‐nothing architecture, meaning that tasks have no dependence on
one other. (This is a slight oversimplification, since the output from mappers is fed to the reducers, but this is under
the control of the MapReduce system; in this case, it needs to take more care rerunning a failed reducer than
rerunning a failed map, since it has to make sure it can retrieve the necessary map outputs, and if not, regenerate
them by running the relevant maps again.) So from the programmer's point of view, the order in which the tasks run
Vidyalankar Institute of Technology Page 68
doesn't matter. By contrast, MPI programs have to explicitly manage their own checkpointing and recovery, which
gives more control to the programmer, but makes them more difficult to write.
MapReduce might sound like quite a restrictive programming model, and in a sense it is: you are limited to key and
value types that are related in specified ways, and mappers and reducers run with very limited coordination between
one another (the mappers pass keys and values to reducers). A natural question to ask is: can you do anything useful
or nontrivial with it
The answer is yes. MapReduce was invented by engineers at Google as a system for building production search
indexes because they found themselves solving the same problem over and over again (and MapReduce was
inspired by older ideas from the functional programming, distributed computing, and database communities), but it
has since been used for many other applications in many other industries. It is pleasantly surprising to see the range
of algorithms that can be expressed in MapReduce, from image analysis, to graph‐based problems, to machine
learning algorithms."" It can't solve every problem, of course, but it is a general data‐processing tool.
When people first hear about Hadoop and MapReduce, they often ask, "How is it different from SETIhome" SETI,
the Search for Extra‐Terrestrial Intelligence, runs a project called SETIhome
in which volunteers donate CPU time
from their otherwise idle computers to analyze radio telescope data for signs of intelligent life outside earth.
SETIhome is the most well‐known of many volunteer computing projects; others include the Great Internet
Mersenne Prime Search (to search for large prime numbers) and Foldinghome (to understand protein folding, and
how it relates to disease).
Volunteer computing projects work by breaking the problem they are trying to solve into chunks called work units,
which are sent to computers around the world to be analyzed. For example, a SETIhome work unit is about 0.35
MB of radio telescope data, and takes hours or days to analyze on a typical home computer. When the analysis is
completed, the results are sent back to the server, and the client gets another work unit. As a precaution to combat
cheating, each work unit is sent to three different machines, and needs at least two results to agree to be accepted.
Although SETIhome may be superficially similar to MapReduce (breaking a problem into independent pieces to be
worked on in parallel), there are some significant differences. The SETIhome problem is very CPU‐intensive, which
makes it suitable for running on hundreds of thousands of computers across the world, " since the time to transfer
the work unit is dwarfed by the time to run the computation on it. Volunteers are donating CPU cycles, not
MapReduce is designed to run jobs that last minutes or hours on trusted, dedicated hardware running in a single
data center with very high aggregate bandwidth interconnects. By contrast, SETIhome runs a perpetual
computation on untrusted machines on the Internet with highly variable connection speeds and no data locality.
A Brief History of Hadoop
Hadoop was created by Doug Cutting, the creator of Apache Lucene, the widely used text search library. Hadoop has
its origins in Apache Nutch, an open source web search engine, itself a part of the Lucene project.
The Origin of the Name "Hadoop"
The name Hadoop is not an acronym; it's a made‐up name. The project's creator, Doug Cutting, explains how the
name came about:
The name my kid gave a stuffed yellow elephant. Short, relatively easy to spell and pronounce, meaningless, and not
used elsewhere: those are my naming criteria. Kids are good at generating such. Googol is a kid's term.
Subprojects and "contrib" modules in Hadoop also tend to have names that are unrelated to their function, often
with an elephant or other animal theme ("Pig," for example). Smaller components are given more descriptive (and
Vidyalankar Institute of Technology Page 69
therefore more mundane) names. This is a good principle, as it means you can generally work out what something
does from its name. For example, the jobtracker5 keeps track of MapReduce jobs.
Building a web search engine from scratch was an ambitious goal, for not only is the software required to crawl and
index websites complex to write, but it is also a challenge to run without a dedicated operations team, since there
are so many moving parts. It's expensive too: Mike Cafarella and Doug Cutting estimated a system supporting a 1‐
billion‐page index would cost around half a million dollars in hardware, with a monthly running cost of 30,000."
Nevertheless, they believed it was a worthy goal, as it would open up and ultimately democratize search engine
Nutch was started in 2002, and a working crawler and search system quickly emerged. However, they realized that
their architecture wouldn't scale to the billions of pages on the Web. Help was at hand with the publication of a
paper in 2003 that described the architecture of Google's distributed filesystem, called GFS, which was being used in
production at Google. GFS, or something like it, would solve their storage needs for the very large files generated as
a part of the web crawl and indexing process. In particular, GFS would free up time being spent on administrative
tasks such as managing storage nodes. In 2004, they set about writing an open source implementation, the Nutch
Distributed Filesystem (NDFS).
In 2004, Google published the paper that introduced MapReduce to the world. Early in 2005, the Nutch developers
had a working MapReduce implementation in Nutch, and by the middle of that year all the major Nutch algorithms
had been ported to run using MapReduce and NDFS.
NDFS and the MapReduce implementation in Nutch were applicable beyond the realm of search, and in February
2006 they moved out of Nutch to form an independent subproject of Lucene called Hadoop. At around the same
time, Doug Cutting joined Yahoo, which provided a dedicated team and the resources to turn Hadoop into a system
that ran at web scale (see sidebar). This was demonstrated in February 2008 when Yahoo announced that its
production search index was being generated by a 10,000‐core Hadoop cluster.
In January 2008, Hadoop was made its own top‐level project at Apache, confirming its success and its diverse, active
community. By this time Hadoop was being used by many other companies besides Yahoo, such as Last.fm,
In one well‐publicized feat, the New York Times used Amazon's EC2 compute cloud to crunch through four terabytes
of scanned archives from the paper converting them to PDFs for the Web.t The processing took less than 24 hours to
run using 100 machines, and the project probably wouldn't have been embarked on without the combination of
Amazon's pay‐by‐the‐hour model (which allowed the NYT to access a large number of machines for a short period),
and Hadoop's easy‐to‐use parallel programming model.
In April 2008, Hadoop broke a world record to become the fastest system to sort a terabyte of data. Running on a
910‐node cluster, Hadoop sorted one terabyte in 209 seconds (just under 3 minutes), beating the previous year's
winner of 297 seconds (described in detail in "TeraByte Sort on Apache Hadoop" on page 461). In November of the
same year, Google reported that its MapReduce implementation sorted one terabyte in 68 seconds.
Hadoop at Yahoo
Building Internet‐scale search engines requires huge amounts of data and therefore large numbers of machines to
process it. Yahoo Search consists of four primary components: the Crawler, which downloads pages from web
servers; the WebMap, which builds a graph of the known Web; the Indexer, which builds a reverse index to the best
pages; and the Runtime, which answers users' queries. The WebMap is a graph that consists of roughly 1 trillion
(10 ) edges each representing a web link and 100 billion (10 ) nodes each representing distinct URLs. Creating and
analyzing such a large graph requires a large number of computers running for many days.
Vidyalankar Institute of Technology Page 70
In early 2005, the infrastructure for the WebMap, named Dreadnaught, needed to be redesigned to scale up to more
nodes. Dreadnaught had successfully scaled from 20 to 600 nodes, but required a complete redesign to scale up
further. Dreadnaught is similar to MapReduce in many ways, but provides more flexibility and less structure. In
particular, each fragment in a Dreadnaught job can send output to each of the fragments in the next stage of the job,
but the sort was all done in library code. In practice, most of the WebMap phases were pairs that corresponded to
MapReduce. Therefore, the WebMap applications would not require extensive refactoring to fit into MapReduce.
Eric Baldeschwieler (Eric14) created a small team and we starting designing and prototyping a new framework
written in C++ modeled after GFS and MapReduce to replace Dreadnaught. Although the immediate need was for a
new framework for WebMap, it was clear that standardization of the batch platform across Yahoo Search was
critical and by making the framework general enough to support other users, we could better leverage investment in
the new platform.
At the same time, we were watching Hadoop, which was part of Nutch, and its progress. In January 2006, Yahoo
hired Doug Cutting, and a month later we decided to abandon our prototype and adopt Hadoop. The advantage of
Hadoop over our prototype and design was that it was already working with a real application (Nutch) on 20 nodes.
That allowed us to bring up a research cluster two months later and start helping real customers use the new
framework much sooner than we could have otherwise. Another advantage, of course, was that since Hadoop was
already open source, it was easier (although far from easy) to get permission from Yahoo's legal department to
work in open source. So we set up a 200‐node cluster for the researchers in early 2006 and put the WebMap
conversion plans on hold while we supported and improved Hadoop for the research users.
The Apache Hadoop Project
Today, T Hadoop is a collection of related subprojects that fall under the umbrella of infrastructure for distributed
computing. These projects are hosted by the Apache Software Foundation, which provides support for a community
of open source software projects. Although Hadoop is best known for MapReduce and its distributed filesystem
(HDFS, renamed from NDFS), the other subprojects provide complementary services, or build on the core to add
higher‐level abstractions. The subprojects, and where they sit in the technology stack, are shown in Figure 1‐1 and
described briefly here:
A set of components and interfaces for distributed filesystems and general I/O (serialization, Java RPC, persistent data
A data serialization system for efficient, cross‐language RPC, and persistent data storage. (At the time of this writing,
Avro had been created only as a new subpro‐ject, and no other Hadoop subprojects were using it yet.)
A distributed data processing model and execution environment that runs on large clusters of commodity machines.
A distributed filesystem that runs on large clusters of commodity machines.
A data flow language and execution environment for exploring very large datasets. Pig runs on HDFS and
A distributed, column‐oriented database. HBase uses HDFS for its underlying storage, and supports both batch‐
style computations using MapReduce and point queries (random reads).
A distributed, highly available coordination service. ZooKeeper provides primitives such as distributed locks
that can be used for building distributed applications.
A distributed data warehouse. Hive manages data stored in HDFS and provides a query language based on SQL
(and which is translated by the runtime engine to MapReduce jobs) for querying the data.
Vidyalankar Institute of Technology Page 71
A distributed data collection and analysis system. Chukwa runs collectors that store data in HDFS, and it uses
MapReduce to produce reports. (At the time of this writing, Chukwa had only recently graduated from a
"contrib" module in Core to its own subproject.).
Vidyalankar Institute of Technology Page 72
When a dataset outgrows the storage capacity of a single physical machine, it becomes necessary to partition it
across a number of separate machines. Filesystems that manage the storage across a network of machines are
called distributed filesystems. Since they are network‐based, all the complications of network programming kick in,
thus making distributed filesystems more complex than regular disk filesystems. For example, one of the biggest
challenges is making the filesystem tolerate node failure without suffering data loss.
Hadoop comes with a distributed filesystem called HDFS, which stands for Hadoop Distributed Filesystem. (You may
sometimes see references to "DFS"—informally or in older documentation or configuration—which is the same
thing.) HDFS is Hadoop's flagship filesystem and is the focus of this chapter, but Hadoop actually has a general‐
purpose filesystem abstraction, so we'll see along the way how Hadoop integrates with other storage systems (such
as the local filesystem and Amazon S3).
The Design of HDFS
HDFS is a filesystem designed for storing very large files with streaming data access patterns, running on clusters on
commodity hardware. Let's examine this statement in more detail:
Very large files
"Very large" in this context means files that are hundreds of megabytes, gigabytes, or terabytes in size. There are
Hadoop clusters running today that store petabytes of data.
Streaming data access
HDFS is built around the idea that the most efficient data processing pattern is a write‐once, read‐many‐times
pattern. A dataset is typically generated or copied from source, then various analyses are performed on that dataset
over time. Each analysis will involve a large proportion, if not all, of the dataset, so the time to read the whole
dataset is more important than the latency in reading the first record.
Hadoop doesn't require expensive, highly reliable hardware to run on. It's designed to run on clusters of commodity
hardware (commonly available hardware available from multiple vendors) for which the chance of node failure
across the cluster is high, at least for large clusters. HDFS is designed to carry on working without a noticeable
interruption to the user in the face of such failure. It is also worth examining the applications for which using HDFS
does not work so well. While this may change in the future, these are areas where HDFS is not a good fit today:
Low‐latency data access
Applications that require low‐latency access to data, in the tens of milliseconds range, will not work well with HDFS.
Remember HDFS is optimized for delivering a high throughput of data, and this may be at the expense of latency.
HBase is currently a better choice for low‐latency access.
Lots of small files
Since the namenode holds filesystem metadata in memory, the limit to the number of files in a filesystem is
governed by the amount of memory on the namenode. As a rule of thumb, each file, directory, and block takes
Vidyalankar Institute of Technology Page 73
about 150 bytes. So, for example, if you had one million files, each taking one block, you would need at least 300 MB
of memory. While storing millions of files is feasible, billions is beyond the capability of current hardware.
Multiple writers, arbitrary file modifications
Files in HDFS may be written to by a single writer. Writes are always made at the end of the file. There is no
support for multiple writers, or for modifications at arbitrary offsets in the file. (These might be supported in
the future, but they are likely to be relatively inefficient.)
HDFS Concepts Blocks
A disk has a block size, which is the minimum amount of data that it can read or write. Filesystems for a
single disk build on this by dealing with data in blocks, which are an integral multiple of the disk block size.
Filesystem blocks are typically a few kilobytes in size, while disk blocks are normally 512 bytes. This is
generally transparent to the filesystem user who is simply reading or writing a file—of whatever length.
However, there are tools to do with filesystem maintenance, such as df and fsck, that operate on the
filesystem block level.
HDFS too has the concept of a block, but it is a much larger unit—64 MB by default. Like in a filesystem for a
single disk, files in HDFS are broken into block‐sized chunks, which are stored as independent units. Unlike a
filesystem for a single disk, a file in HDFS that is smaller than a single block does not occupy a full block's
worth of underlying storage. When unqualified, the term "block" in this book refers to a block in HDFS.
Why Is a Block in HDFS So Large
HDFS blocks are large compared to disk blocks, and the reason is to minimize the cost of seeks. By making a
block large enough, the time to transfer the data from the disk can be made to be significantly larger than
the time to seek to the start of the block. Thus the time to transfer a large file made of multiple blocks
operates at the disk transfer rate.
A quick calculation shows that if the seek time is around 10ms, and the transfer rate is 100 MB/s, then to
make the seek time 1 of the transfer time, we need to make the block size around 100 MB. The default is
actually 64 MB, although many HDFS installations use 128 MB blocks. This figure will continue to be revised
upward as transfer speeds grow with new generations of disk drives.
This argument shouldn't be taken too far, however. Map tasks in MapReduce normally operate on one block
at a time, so if you have too few tasks (fewer than nodes in the cluster), your jobs will run slower than they
Having a block abstraction for a distributed filesystem brings several benefits. The first benefit is the most
obvious: a file can be larger than any single disk in the network. There's nothing that requires the blocks
from a file to be stored on the same disk, so they can take advantage of any of the disks in the cluster. In
fact, it would be possible, if unusual, to store a single file on an HDFS cluster whose blocks filled all the disks
in the cluster.
Second, making the unit of abstraction a block rather than a file simplifies the storage subsystem. Simplicity
is something to strive for all in all systems, but is important for a distributed system in which the failure
modes are so varied. The storage subsystem deals with blocks, simplifying storage management (since
blocks are a fixed size, it is easy to calculate how many can be stored on a given disk), and eliminating
Vidyalankar Institute of Technology Page 74
metadata concerns (blocks are just a chunk of data to be stored—file metadata such as permissions
information does not need to be stored with the blocks, so another system can handle metadata
Furthermore, blocks fit well with replication for providing fault tolerance and availability. To insure against
corrupted blocks and disk and machine failure, each block is replicated to a small number of physically
separate machines (typically three). If a block becomes unavailable, a copy can be read from another
location in a way that is transparent to the client. A block that is no longer available due to corruption or
machine failure can be replicated from their alternative locations to other live machines to bring the
replication factor back to the normal level. (See "Data Integrity" on page 75 for more on guarding against
corrupt data.) Similarly, some applications may choose to set a high replication factor for the blocks in a
popular file to spread the read load on the cluster.
Like its disk filesystem cousin, HDFS's fsck command understands blocks. For example, running:
hadoop fsck ‐files ‐blocks
will list the blocks that make up each file in the filesystem.
Namenodes and Datanodes
A HDFS cluster has two types of node operating in a master‐worker pattern: a name‐node (the master) and a
number of datanodes (workers). The namenode manages the filesystem namespace. It maintains the
filesystem tree and the metadata for all the files and directories in the tree. This information is stored
persistently on the local disk in the form of two files: the namespace image and the edit log. The namenode
also knows the datanodes on which all the blocks for a given file are located, however, it does not store
block locations persistently, since this information is reconstructed from datanodes when the system starts.
A client accesses the filesystem on behalf of the user by communicating with the name‐node and datanodes.
The client presents a POSIX‐like filesystem interface, so the user code does not need to know about the
namenode and datanode to function.
Datanodes are the work horses of the filesystem. They store and retrieve blocks when they are told to (by
clients or the namenode), and they report back to the namenode periodically with lists of blocks that they
Without the namenode, the filesystem cannot be used. In fact, if the machine running the namenode were
obliterated, all the files on the filesystem would be lost since there would be no way of knowing how to
reconstruct the files from the blocks on the datanodes. For this reason, it is important to make the
namenode resilient to failure, and Hadoop provides two mechanisms for this.
The first way is to back up the files that make up the persistent state of the filesystem metadata. Hadoop
can be configured so that the namenode writes its persistent state to multiple filesystems. These writes are
synchronous and atomic. The usual configuration choice is to write to local disk as well as a remote NFS
Vidyalankar Institute of Technology Page 75
It is also possible to run a secondary namenode, which despite its name does not act as a namenode. Its main role is
to periodically merge the namespace image with the edit log to prevent the edit log from becoming too large. The
secondary namenode usually runs on a separate physical machine, since it requires plenty of CPU and as much
memory as the namenode to perform the merge. It keeps a copy of the merged namespace image, which can be
used in the event of the namenode failing. However, the state of the secondary namenode lags that of the primary,
so in the event of total failure of the primary data, loss is almost guaranteed. The usual course of action in this case
is to copy the namenode's metadata files that are on NFS to the secondary and run it as the new primary.
The CommandLine Interface
We're going to have a look at HDFS by interacting with it from the command line. There are many other interfaces to
HDFS, but the command line is one of the simplest, and to many developers the most familiar.
We are going to run HDFS on one machine, so first follow the instructions for setting up Hadoop in pseudo‐
distributed mode in Appendix A. Later you'll see how to run on a cluster of machines to give us scalability and fault
There are two properties that we set in the pseudo‐distributed configuration that deserve further explanation. The
first is fs.default.name, set to hdfs://localhost/, which is used to set a default filesystem for Hadoop. Filesystems are
specified by a URI, and here we have used a hdfs URI to configure Hadoop to use HDFS by default. The HDFS
daemons will use this property to determine the host and port for the HDFS namenode. We'll be running it on
localhost, on the default HDFS port, 8020. And HDFS clients will use this property to work out where the namenode
is running so they can connect to it.
We set the second property, dfs.replication, to one so that HDFS doesn't replicate filesystem blocks by the usual
default of three. When running with a single datanode, HDFS can't replicate blocks to three datanodes, so it would
perpetually warn about blocks being under‐replicated. This setting solves that problem.
Basic Filesystem Operations
The filesystem is ready to be used, and we can do all of the usual filesystem operations such as reading files, creating
directories, moving files, deleting data, and listing directories. You can type hadoop fs ‐help to get detailed help on
every command. Start by copying a file from the local filesystem to HDFS:
hadoop fs ‐copyFromLocal input/docs/quangle.txt hdfs://localhost/user/tom/quangle.txt
This command invokes Hadoop's filesystem shell command fs, which supports a number of subcommands—in this
case, we are running ‐copyFromLocal. The local file quangle.txt is copied to the file /user/tom/quangle.txt on the
HDFS instance running on localhost. In fact, we could have omitted the scheme and host of the URI and picked up
the default, hdfs://localhost, as specified in core‐site.xml.
hadoop fs ‐copyFromLocal input/docs/quangle.txt /user/tom/quangle.txt
We could also have used a relative path, and copied the file to our home directory in HDFS, which in this case is
hadoop fs ‐copyFromLocal input/docs/quangle.txt quangle.txt
Let's copy the file back to the local filesystem and check whether it's the same:
hadoop fs ‐copyToLocal quangle.txt quangle.copy.txt md5 input/docs/quangle.txt quangle.copy.txt
Vidyalankar Institute of Technology Page 76
MD5 (input/docs/quangle.txt) = al6f23lda6b05e2ba7a339320e7dacd9 MD5 (quangle.copy.txt) =
The MD5 digests are the same, showing that the file survived its trip to HDFS and is back intact. Finally, let's look at
an HDFS file listing. We create a directory first just to see how it is displayed in the listing:
hadoop fs ‐mkdir books hadoop fs ‐ls .
Found 2 items
drwxr‐xr‐x ‐ tom supergroup 0 2009‐04‐02 22:41 /user/tom/books
‐rw‐r‐‐r‐‐ 1 tom supergroup 118 2009‐04‐02 22:29 /user/tom/quangle.txt
The information returned is very similar to the Unix command ls ‐l, with a few minor differences. The first column
shows the file mode. The second column is the replication factor of the file (something a traditional Unix filesystems
does not have). Remember we set the default replication factor in the site‐wide configuration to be 1, which is why
we see the same value here. The entry in this column is empty for directories since the concept of replication does
not apply to them—directories are treated as metadata and stored by the namenode, not the datanodes. The third
and fourth columns show the file owner and group. The fifth column is the size of the file in bytes, or zero for direc‐
tories. The six and seventh columns are the last modified date and time. Finally, the eighth column is the absolute
name of the file or directory.
File Permissions in HDFS
HDFS has a permissions model for files and directories that is much like POSIX. There are three types of permission:
the read permission (r), the write permission (w) and the execute permission (x). The read permission is required to
read files or list the contents of a directory.
The write permission is required to write a file, or for a directory, to create or delete files or directories in it.
The execute permission is ignored for a file since you can't execute a file on HDFS (unlike POSIX), and for a directory
it is required to access its children.
Each file and directory has an owner, a group, and a mode. The mode is made up of the permissions for the user who
is the owner, the permissions for the users who are members of the group, and the permissions for users who are
neither the owner nor members of the group.
A client's identity is determined by the username and groups of the process it is running in. Because clients are
remote, this makes it possible to become an arbitrary user, simply by creating an account of that name on the
Thus, permissions should be used only in a cooperative community of users, as a mechanism for sharing filesystem
resources and for avoiding accidental data loss, and not for securing resources in a hostile environment. However,
despite these drawbacks, it is worthwhile having permissions enabled (as it is by default; see the dfs.permissions
property), to avoid accidental modification or deletion of substantial parts of the filesystem, either by users or by
automated tools or programs.
When permissions checking is enabled, the owner permissions are checked if the client's username matches the
owner, and the group permissions are checked if the client is a member of the group; otherwise, the other
permissions are checked.
Vidyalankar Institute of Technology Page 77
There is a concept of a super‐user, which is the identity of the namenode process. Permissions checks are not
performed for the super‐user.
Hadoop has an abstract notion of filesystem, of which HDFS is just one implementation. The Java abstract class
org.apache.hadoop.fs. FileSystem represents a filesystem in Hadoop, and there are several concrete
implementations, which are described below:
Filesystem URI Java implementation (all Description
scheme under org.apache.hadoop)
Local file fs.LocalFileSystem A filesystem for a locally connected disk with client‐side check‐
sums. Use RawLocalFileSys tem for a local filesystem with no
checksums. See "LocalFileSys‐tem" on page 76.
HDFS hdfs hdfs.DistributedFileSystem Hadoop's distributed filesystem. HDFS is designed to work effi‐
ciently in conjunction with Map‐Reduce.
HFTP hftp hdfs.HftpFileSystem A filesystem providing read‐only access to HDFS over HTTP.
(Despite its name, HFTP has no connection with FTP.) Often used
with distcp ("Parallel Copying with distcp" on page 70) to copy
data between HDFS clusters running different versions.
HSFTP hsftp hdfs.HsftpFileSystem A filesystem providing read‐only access to HDFS over HTTPS.
(Again, this has no connection with FTP.)
HAR har fs.HarFileSystem A filesystem layered on another filesystem for archiving files. Ha‐
doop Archives are typically used for archiving files in HDFS to
reduce the namenode's memory usage.
KFS (Cloud‐ kfs fs.kfs.KosmosFileSystem CloudStore (formerly Kosmos filesystem) is a distributed filesys‐
Store) tem like HDFS or Google's GFS, written in C++. Find more infor‐
mation about it at http://kosmosfs .sourceforge.net/.
FTP ftp fs.ftp.FTPFileSystem A filesystem backed by an FTP server.
S3 (native) s3n fs.s3native.NativeS3FileSys A filesystem backed by Amazon S3. See http://wiki.apache.org/ha
S3 (block‐ s3 fs.s3.S3FileSystem A filesystem backed by Amazon S3, which stores filesin blocks
Note: Don’t go too much into the technical aspects of Hadoop HDFS. You need to understand the concept and
derive from that an understanding of unlocking old data. Fundamentally these technologies allow for mining data
that otherwise could not be mined, simply by supporting distributed computing and storage. Also, use the slides
to figure out how a MapReduce program works and the various entities involved.
Vidyalankar Institute of Technology Page 78