Transcript Document

IS 483
Information Systems Management
James Nowotarski
17 April 2003
Today’s Objectives
• Recap the IT-Human Resources Management
function
• Build awareness of distributed
architectures/infrastructures
• Build awareness of network management
• Review logistics for Assignments 1, 2, and 3
Today’s agenda
Topic
Duration
• Recap and update IT HRM
20 minutes
• Distributed architectures
55 minutes
*** Break
15 minutes
• Distributed architectures (cont.)
50 minutes
• Quiz
25 minutes
• Assignments 2,3
15 minutes
Today’s agenda
Topic
Duration
• Recap and update IT HRM
20 minutes
• Distributed architectures
55 minutes
*** Break
15 minutes
• Distributed architectures (cont.)
50 minutes
• Quiz
25 minutes
• Assignments 2,3
15 minutes
Centralized vs. Decentralized
organizational models
The Federal model results in the optimal balance
between centralized and decentralized
Federal model
Shared vision &
leadership
Unresponsive
No BU ownership
Doesn’t meet every
BU’s needs
Lacks content depth
Purely centralized
Consistent
quality
Economies
of scale
Shared
standards &
infrastructure
Speed
Access to MU
content experts
Responsive to
BU’s needs
Critical mass
Synergy
of skills
&
Integration
Mutual trust &
commitment
Redundant
functions/costs
Proliferation of assets,
delivery vehicles, rollouts
Variable quality
Lack of synergy and
integration
Purely decentralized
Source: MIT
Flamholtz Leadership Effectiveness
Framework
Situational Factors
Work
Style
Effectiveness
People
Task
Situation determines the style that will most likely be effective
Factors Affecting
Optimal Leadership Style
Optimal leadership style depends on degree of
job autonomy and task programmability
Potential for Job Autonomy
Low
High
High
Directive
Interactive
Low
Interactive
Nondirective
Task Programmability
IT Human Resource
Management Components
1.
2.
3.
4.
5.
6.
Hiring
Turnover and Retention
Career Development/Path
Training
Compensation
Performance Appraisal
Role of Advanced Automation in HR
• Cross-functional services are possible- such as
merging payroll, pension, and benefits information
for a single employee
• Training and employee self-service can become a
part of the Human Resource system
• Expert Systems available that answer questions
typically answered by Human Resource
professionals (e.g., HR policy questions)
• Use of Internet for recruiting, benchmarking, and
anywhere/anytime access
Source: Groe, Gerald M; Pyle, William; Jamrog, Jay J. “Information Technology and HR.”
Human Resource Planning. 1996. Vol. 19, number 1.
http://facweb.cs.depaul.edu/nsutcliffe/483readings/Groe-IT%20and%20HR-session3.htm
Today’s agenda
Topic
Duration
• Recap and update IT HRM
20 minutes
• Distributed architectures
55 minutes
*** Break
15 minutes
• Distributed architectures (cont.)
50 minutes
• Quiz
25 minutes
• Assignments 2,3
15 minutes
Solution “Stack”
Examples
Applications and
Data
Reusable Technical
Architecture
System Software
Hardware/Network
Claims, Payroll, CRM, ERP
Component-Based
Architecture for
Microsoft (Accenture)
Middleware
WebSphere, CICS TS
Apache
J2EE
DB2, Oracle, SQL Server
Linux, Unix, Windows, z/OS
Dell, HP, Sun, EMC, Cisco
AT&T, Sprint, WorldCom
Public Internet
IT Architecture and IT Infrastructure
Applications and
Data
Reusable Technical
Architecture
IT Architecture
System Software
IT Infrastructure
Hardware/Network
IT Architecture vs. IT Infrastructure
IT Architecture
Level of
Abstraction
Scope
• Blueprint
• Policies, rules
• Tech architecture
• System software
• Hardware/Network
IT Infrastructure
Implemented capabilities
IT Architecture
+ IT roles, organization
+ IT processes
IT Infrastructure
in·fra·struc·ture n. 1. An underlying base or foundation,
especially for an organization or a system.
2. The basic facilities, services, and installations needed for
the functioning of a community or society, such as
transportation and communication systems, water and power
lines, and public institutions including schools, post offices,
and prisons.
Source: IBM
IT Architecture
arch·i·tec·ture n. 1. A formal specification for how a computer
solution will be organized and how its parts inter-relate
2. An integrated set of technology choices used to guide the
organization
Sources: Gartner Group, MIT
Benefits of Architecture
•
•
•
•
•
•
•
•
Productivity
Consistency
Quality
Rapid delivery
Maintainability
Interoperability
Reduces complexity
Leverages scarce skills
Applications and Data
Applications and
Data
Reusable Technical
Architecture
System Software
Hardware/Network
• Presentation logic
• Application logic
• Data management
What is Distributed Computing?
More than 1 interconnected processor
operating at the same time
Five Types of Distributed Computing
Distributed
Presentation
Remote
Presentation
Distributed
Function
Remote Data
Management
Data
Management
Data
Management
Data
Management
Data
Management
Application
Logic
Application
Logic
Application
Logic
Distributed
Database
Data
Management
Presentation
Data
Management
Presentation
Presentation
Application
Logic
Application
Logic
Application
Logic
Presentation
Presentation
Presentation
Apps &
Data
Source: Gartner Group
Where Does Internet Fit?
Distributed
Presentation
Remote
Presentation
Distributed
Function
Remote Data
Management
Data
Management
Data
Management
Data
Management
Data
Management
Application
Logic
Application
Logic
Application
Logic
Distributed
Database
Data
Management
Presentation
Data
Management
Presentation
Thin
Presentation
Application
Logic
Application
Logic
Application
Logic
Presentation
Presentation
Presentation
Thick
Thickness
of Client
Distributed vs. Internet
Internet = Distributed + Reach + Content
(Netcentric)
(Client/Server)
Other differences:
• Internet applications are more dynamic, less static (thin vs. thick client)
• Internet applications offer more standard user interface
2-Tier Distributed Architecture
“Tiers” are different types or groupings of
computers used to perform different functions1
Wide area
network (WAN)
Workstation
• Presentation
• Application logic
1 Above
Enterprise
Server
• Application logic
• Data management
definition focuses on hardware, but tiers may also refer to software
3-Tier Distributed Architecture
LAN
WAN
Workstation
Workgroup
Server
• Presentation
• Application logic
Enterprise
Server
• Data management
Comparison of 3-tier relative to 2-tier
Advantages
• Better load balancing on different servers
• More scalable
Disadvantages
• Potentially greater load on network
• More complex to implement
• More hardware and network components ($$$)
Why is it more complex?
The Fundamental Counting Principle
First event may occur in n1 different ways, the
second event may occur in n2 different ways and so
on, the total number of ways the compound event
may occur is
n1 • n2 • n 3 • . . . . • nk
n-Tier Example
DB
Server
Bank
Customers
Internet
Web
Server
App
Server
Legacy
Mainframe
Internet
Firewall
Application
Firewall
Comparison of n-tier relative to 3-tier
Advantages
• Better load balancing on different servers
• More scalable
Disadvantages
• Potentially greater load on network
• More complex to implement
• More hardware and network components ($$$)
Solution “Stack”
Examples
Applications and
Data
Reusable Technical
Architecture
System Software
Hardware/Network
Claims, Payroll, CRM, ERP
Component-Based
Architecture for
Microsoft (Accenture)
Middleware
overlap
WebSphere, CICS TS
Apache
J2EE
DB2, Oracle, SQL Server
Linux, Unix, Windows, z/OS
Dell, HP, Sun, EMC, Cisco
AT&T, Sprint, WorldCom
Public Internet
Middleware
Middleware is a layer of software
between applications/technical
architecture and systems
software/network
Functions
• Typically provides high-level application
programming interfaces (APIs) to shields
programmers from complex protocols
• Enables disparate components to communicate
Middleware
Example 1
Application
CICS TS
z/OS
Middleware, providing:
• logging
• security
• recovery/restart
• 2-phase commit
Middleware
Example 2
PlatformA
PlatformB
Client
Object
Server
Object
Object Request Broker (ORB)
• Client uses ORB to transparently invoke a function on
a server object
• ORB provides interoperability
• ORB isolates client from location, programming
language, and operating system of server object
Middleware
Types of Middleware
• Network middleware
– Procedure calls over a network
– Example: Remote Procedure Calls (RPCs), ORBs
• Conversational middleware
– Transparent conversion of text, graphics, data
– Example: Electronic Data Interchange (EDI)
• Transaction Processing
– Transaction management services to support processing of
high volumes of transactions
– Example: CICS TS
Middleware
Types of Middleware (cont.)
• SQL and Database Middleware
– Applications can connect to several different databases
using one programming interface
– Example: Open Database Connectivity (ODBC)
• GUI Middleware
– Applications can support multiple GUIs across a network
– Example: X Windows
• Messaging Middleware
– Distributing data and control through synchronous or
asynchronous messages
– Example: IBM MQSeries
IT Management Issues Associated with
Distributed Architectures
•
Migrating from legacy (mainframe) to distributed
architecture:
 3 major business issues
 5 typical alternatives
•
Support and maintenance issues
Business Issues to Consider
When migrating a legacy application to a distributed
architecture, consider cost, resources, and risk
Cost
•
Generally lower costs after migration
• hardware
• software
• depreciation (2-3 years vs. 5-7 years for mainframes)
• development cycle times are shorter
Business Issues to Consider
When migrating a legacy application to a distributed
architecture, consider cost, resources, and risk
Resources
•
•
•
Qualifications of individuals assigned
Time that these individuals spend
Level of system management software available
• Tried and true tested software exists for mainframes but is
spotty for a multi-platform environment where some aspects
are centralized and others are distributed
Business Issues to Consider
When migrating a legacy application to a distributed
architecture, consider cost, resources, and risk
Risk
•
Risks in terms of reliability, availability, serviceability, and security are
managed when the migration is done in 3 phases:
1. Baseline inventory established -- all application elements are
inventoried including source code, data files, test data, and
test results
2. Migration process can now begin
3. Final system test, parallel runs, and production cutover
phase
Migration Alternatives
1. Re-platforming
2. “Screen scraping”
3. Replace legacy with package that is built on
distributed architecture
•
change management issues
4. Migrate from scratch
5. #4 + Business process reengineering
Support/Maintenance Issues
•
Training and standards in more sophisticated
design techniques, testing, system operation, etc.
•
Robust, fault tolerant components may not be
available, may need to prepare to operate in a
degraded mode
•
Service level monitoring and troubleshooting are
more complex and difficult with distributed
architectures
Trends in IT Architecture and
Infrastructure
•
Globalization
•
Real-time enterprise
•
Mobility
•
Open Source
•
Application integration and web services
 Intra-enterprise
 Inter-enterprise
Trends in IT Architecture and
Infrastructure
Competition is fierce
•
“[Larry] Ellison [CEO of Oracle] cited the
success of the Apache Web server over
Microsoft's Internet Information Services (IIS)
technology as an example, declaring that the
Redmond software giant has been ‘slaughtered,
wiped out, taken from market dominance to
irrelevance’ by the open-source
product. ” (Application Development Trends,
4/14/03)
Business Drivers
•
Global competition
•
Profit pressure
•
Market differentiation
•
Responsiveness
•
Link to increased business value
 Delicate balance between stable, yet agile
infrastructure
Today’s agenda
Topic
Duration
• Recap and update IT HRM
20 minutes
• Distributed architectures
55 minutes
*** Break
15 minutes
• Distributed architectures (cont.)
50 minutes
• Quiz
25 minutes
• Assignments 2,3
15 minutes
Today’s agenda
Topic
Duration
• Recap and update IT HRM
20 minutes
• Quiz
25 minutes
• Distributed architectures
30 minutes
*** Break
15 minutes
• Distributed architectures (cont.)
75 minutes
• Assignments 2,3
15 minutes
Topics for April 24
• Network Management
• See web site for reading
• Quiz on Network Management
• Operations Management (start)
End of slides - extra slides follow
Network Management
Definition
The set of activities required to plan, install, monitor and
maintain all network components in order to achieve
specified service levels reliably, at an acceptable
and agreed cost
Importance of business focus
in network solutions
• Organizations need to view the network as critical
to the lifeline of the business and ensure the
network is serving the needs of their business
• A complete solution involves managing the
network and all its components which includes the
business processes and user’s needs
• The solution should also provide a method that
shows the business processes, the business units,
and who will be impacted by the problem
• Administrators will need to proactively manage the
network through policy based procedures instead
of reacting to user’s complaints
Network management architecture
Source: Cisco