Information Systems Planning
1
2 Distributed Systems Architecture
Information Systems Planning
Paradox of IS Planning
Most organization's survival now depends on ITPlanning of its effective use is a matter of organizational life and death
IT is changing so fastIs it useless to do IS planning?
A variety of approaches, tools and mechanisms available for IS planning
No best way to do it.
Mindset for Planning
Traditional viewDetermining what decisions to make in the future
A better viewDeveloping a view of the future that guides decision making today
Difference : strategy making instead of planning
Strategy: stating the direction in which you want to go and how you intend to get there
Types of Planning
Horizon Focus Issues Primary Responsibility
3 - 5 years
Strategic Vision, architecture, business goals
Senior management, CIO
1 - 2 years
Tactical Resource allocation, project selection
Middle management,IS line partners, Steering committees
6 month -1 year
Operational Project management, meeting time and budget targets
IS professionals, Line managers, partners
Why Planning Is So Difficult? (1)
Business goals and systems plans need to alignStrategic systems plans need to align with business goals and support those objectivesWill be difficult if CIO is not part of senior management
Technologies are rapidly changingContinuous planning based on monitoring and experimenting new technologiesAdvanced technology groups
Why Planning Is So Difficult? (2)
Companies need portfolios rather than projects Evaluation on more than their individual meritHow they fit into other projects and how they balance the portfolio of projects
Infrastructure development is difficult to fundOften done under the auspices of a large application projectChallenge: develop improved applications and improve infrastructure over time
Mainframe C/S ERP Web application Web Services
Why Planning Is So Difficult? (3)
Responsibility Needs to be JointSystems planning has become business planning, not just a technology issueIt is better done by a full partnership of C-level officers
Other planning issuesTop-down Vs. bottom-up
Radical change Vs. continuous
Planning culture in which the systems planning must fit
Tradition Strategy Making
Assumptions:The future can be predictedTime is available to do these 3 partsIS supports and follows the businessTop management knows best (broadest view of firm)Company: like an "Army"
Business Strategy•Business decision•Objectives and direction•Change
System Strategy•Business-based•Demand-oriented•Application-focused
IT Strategy•Activity-based•Supply-oriented•Technology-focused
Supportsbusiness
DirectionFor IS
Infrastructure and services
Needs and priorities
Step 1Where is the business going and why?
Step 2What is required?
Step 3How can it be delivered?
A World of Rapid Change (1)
Today, due to the Internet and other technological advances, these assumptions no longer hold true:
The future cannot be predictedDiscontinuous change
Who predicted Internet, Amazon, eBay etc.?
Time is not available for the sequenceNever enough time in Internet AgeIT implementation planning needs to go ahead of business strategizing
A World of Rapid Change (2)
IS does not JUST support the business anymore
Top management may not know best
Inside out Vs. outside in approach
An organization is not like an army
Industrial era metaphor no longer always applies Core
andCustomer
Today's Sense-and-Response Approach (1)
Let strategies unfold rather than plan them:
A sense-and-respond approach when predictions are risky
Sense a new opportunity and immediately respond by testing itMyriad of small experiments
Time
TimeStrategic envelop
Old-era strategyOne big choice, long commitment
New-era strategyMany small choices, short commitments
Case Example: Microsoft
Abandoned proprietary network despite big investment when it did not capture enough customersMoved on to buying Internet companies as well as aligning with Sun to promote JavaOver time, they moved into a variety of technologies:
Web, Cable news, Digital movies, Cable modems, Handheld OS, Video server, Music, Xbox, .Net, Search engines...
Not all strategies came from top managemente.g. first server came from a rebel's unofficial project
Getting its fingers into every pie that might become important
Today's Sense-and-Response Approach (2)
Formulate strategy closest to the action:Close contact with the market
Employees who interact daily with customers, suppliers and partners (organizational edges)
Employees who are closest to the future should become prime strategists.
In the Internet Age, this means younger employees
Today's Sense-and-Response Approach (3)
Guide strategy-making with a “strategic envelope":
Having a myriad of potential corporate strategies being tested in parallel could lead to anarchy without a central guiding mechanismTop management set the parameters for the experiments, and then continually manage that context
Experiment by territory (as Microsoft did)Strategic conversationMeet regularly with the experimenters
Today's Sense-and-Response Approach (4)
Be at the TableIS executives should be actively involved in business strategizingThe IS function needs to be strategy-orientedCIO need to make their departments credible and outsource most operational work
Test the FutureNeed to test potential futures before the business is ready for them (thinking ahead of the business)
Provide funding for experimentsWork with research organizationsHave an emerging technologies group
Today's Sense-and-Response Approach (5)
Put the Infrastructure in Place:Moving quickly in Internet commerce means having the right IT infrastructure in place.IT experiments are recommended to include those that test painful infrastructure issues
Create and maintain common, consistent data definitionsCreate and instill mobile commercial standards among handheld devicesImplement e-commerce security and privacy measuresDetermine operational platforms (ERP, Supply Chain Management …)
Stages of Growth (1)
Richard Nolan et al observed four stages in the introduction and assimilations of a new technology
Early SuccessesIncreased interest and experimentation
ContagionInterest grows rapidly; growth is uncontrolled; learning period for the field
ControlEfforts begun toward cost reduction and standardization
IntegrationDominant design mastered; setting the stage for newer technology
Stages of Growth (2)O
rgan
izat
iona
l Lea
rnin
g
Time1960 1980 1995 2010
DP EraMicro Era
Network Era
Technological discontinuity
Stage 1:Initiation
Stage 2:Contagion
Stage 3:Control
Stage 4:Integration and Stage 1 of Micro Era
Stage 2:Contagion
Stage 3:Control
Stage 4:Integration and Stage 1 of Network Era
Stage 2:Contagion
Stage 3:Control
Stage 4:Integration
Stages of Growth (3)
The importance of the theory is understanding where a technology or company resides on the organizational learning curve
e.g. Web Service is currently in Stage 2, too much control at the learning and experimentation stage can kill off new uses of technology
Management principles differ from stage to stageDifferent technologies are in different stages at any point in time
Competitive Forces Model (1)
Michael Porter's Five Forces ModelA model that determines the relative attractiveness (competition) of an industry.
Five forcesBargaining power of customers and buyers
High when buyers have many choices of whom to buy from, and low when the choices are few.
Bargaining power of suppliersHigh when buyers have few choices of whom to buy from, and low when there are many choices.
Competitive Forces Model (2)
Threat of substitute products or servicesLow if there are very few alternatives to replace the product or service.Switching costs
Costs that can make customers reluctant to switch to another product or service.
Threat of new entrantsHigh when it is easy for competitors to enter the market
The intensity of rivalry among competitorsHigh when the industry is less attractive.
Competitive Forces Model (3)
Industry Competitors
Rivalry among existing firms
Suppliers Customers and Buyers
Potential Entrants
Substitutes
Bargaining power of buyers
Threat of substitute products or services
Bargaining power of suppliers
Threat of new entrants
How will the business react to threats (and opportunities)?
Competitive Forces Model (4)
Industry Competitors
Rivalry among existing firms
Suppliers
Potential Entrants
Substitutes
Bargaining power of buyers
Threat of substitute products or services
Bargaining power of suppliers
Threat of new entrants
Customers and Buyers
The strategy and actions an organization adopts depend upon its perceptions of itself and these threats.
Porter’s strategies:• Product differentiation (non-duplicable product or service)• Low-cost producer • Market niche (market segment or geographical market)
Five Forces Analysis of the Internet
The Internet tends to dampen the profitability of industries
Increases the bargaining power of buyersDecreases barriers to entryIncreases the bargaining power of suppliersIncreases the threat of substitute products and servicesIntensifies rivalry among competitors
Success depends on offering distinct valueFirms should focus on their strategic position in an industry and how they will maintain profitability
1
2 Distributed Systems Architecture
Information Systems Planning
Definition: IT Architecture VS. IT Infrastructure
An IT architecture is a blueprint showing how the parts will interact and interrelate.
System, information, departments...Multiplicity of structures and views
An IT infrastructure is the implementation of an architecture.
processors, software, databases, electronic links, data centers, standards, skills, electronic processes...We now tend to divide computing into applications and infrastructures
The Evolution of Distributed Systems
Mainframes: with dumb terminalsMinicomputers moved computing into departments
The master-slave computing model persisted and processing was mainly centralized
Microcomputer moved processing power into desktop, briefcase and handhelds
Client/server computing
Internet: a globally distributed systemInteresting twist: power returning to a type of centralized processing with networks of servers
Four Attributes of Distributed Systems
The degree to which a system is distributed can be determined by answering four questions:
1. Where is the processing done? 2. How are the processors and other devices
interconnected?3. Where is the information stored?4. What rules or standards are used?
Distributed Processing
Limited processing power of a single node VS. increasing application demands
Balance the load and improve overall performanceLet machines handle the work they do best
Interoperability: information exchange between heterogeneous computing platforms
ProtocolsTwo-way message passing between user applications
Connectivity Among Processors
Data exchanges through electronic communication links
TCP/IPEthernet, ATM, FDDI, Frame relay...
Planned Redundancy for reliabilityTwo or more independent paths between two nodes to provide automatic alternate routingTopology and reliability of the Internet
Distributed Databases
Two distributed database schemesDivide a database and distribute its portions throughout a system without duplicating the data
Transparent user access
Store the same data at several different locations, with one site containing the master file
Synchronization issueE.g. edge servers
System-wide Rules
Rules governing communication between nodes, security, data accessibility, program and file transfers, and common operating procedures
Open standards after 1990sOSI Reference ModelSQLAPI: standardized interfaceTCP/IPOpen source
Internet---A Scale-free Network (1)
Internet is not designed, but evolvedInternet is a scale-free network
Scale-free networks are very common and a very important category of real networks.
Scale-free networks are the direct result of self-organized growth
Growth: networks continuously expand by the addition of new nodesSpecial type of growth called preferential attachment
Preferential Attachment : The attachment is NOT uniform A node is linked with higher probability to a node that already has a large number of links
Internet---A Scale-free Network (2)
Five nodes with most linksFirst neighbors of red nodes
27% reach
60% reach
Random/
Internet---A Scale-free Network (3)Poisson distribution
Exponential Network
Power-law distribution
Scale-free Network
P(k)~k-γ
Internet---Topological Robustness
Topological robustness: the Internet is robust in the presence of random failures.
At any given time hundreds of routers are down but the performance is not impactedIt will function even if we remove randomly 80% of the nodes.
Theoretical and experimental investigations show that scale-free networks are topologically robust
Internet---Vulnerability to Targeted Attacks
Scale-free networks such as Internet are vulnerable to attacks.
If a malicious attack could simultaneously remove 5% of hubs (the highly connected nodes) the network would disintegrate
Internet---Virus
Scale-free networks like internet are vulnerable to spreading viruses
Hubs are passing them massively to the connected multiple nodes.
This suggests immunizing hubs.
When to Distribute Computing Responsibilities (1)
The decision of distributing computing responsibilities is rather managerial than technical
People deciding how their portion of business operates should also decide how they use IT
When to Distribute Computing Responsibilities (2)
Systems responsibilities can be distributed unless the following are true:
Are the operations interdependent? For interdependent, their planning, development, resources, and operations must be centrally coordinated
Are the businesses really homogenous? Processing may be distributed, but planning and hardware selection should be centralized
Does the corporate culture support decentralization? Corporate culture might centralize finance, HR, and systems planning
An Organizational Framework
Systems may be needed for all 7 levelsInter-organizational links can occur at all six internal levels
The current hot levelsLevel1: inter-enterprise computingLeve5: where business processes resideIndividual
Work Group or Team
Department or Process
Plant or Site
Country or Region
Enterprise
Business Ecosystem1
2
3
4
5
6
7
A Technical Framework (1)
The SUMURU architecture developed in 1982, has stood the test of time.
It provides a clear conceptual framework for understanding various components of a distributed architecture.
A Technical Framework (2)
A Technical Framework (3)
Processors Services
Single-user systems (SU) Terminal access
Multiple-user systems (MU) File transfer
Remote utility systems (RU) Computer mail
Networks Standards
Local network (LN) Operating system
Remote network (RN) Communications protocols
Database systems