FALL ‘98 FORUM

September 18, 1998

Sheraton Denver West - 360 Union Boulevard, Lakewood CO (303) 987-2000

AGENDA

8:00-8:30 Registration

8:30-9:00 Annual Business Meeting

9:00-10:00 "DB2 Performance Analysis For Mainframe Specialists"
Ned A. Diehl, The Information Systems Manager

DB2 is a complex DBMS that has become a significant workload delivery facility for more large system environments. As with many other subsystems, available performance data is different than what traditional large system performance analysts and capacity planners have become accustomed to. However, like other delivery vehicles, DB2 uses CPU cycles, memory and DASD I/O devices. This paper is intended for large system performance analysts and focuses on data that can be used for traditional analysis. Detailed DB2 knowledge is not required.

10:00-10:15 Break

10:15-11:30 "The Folly of Ownership, Revisited or Whose CICS Region is that, Anyway?"
Steve Samson, Candle Corporation

A nearly universal trend in large IT installations has been to transform resources "owned" by user departments to installation ownership with more or less automated management. Examples include DASD volumes and JES initiators. However, one counter example to this trend persists: CICS address spaces are still owned overwhelmingly by user departments or the proxies who maintain their inventory of applications. Consequently, the set of transactions resident in each region mixes the important and the unimportant, the sprinter and the slug. Since the execution environment is managed at the address space level the effect of such transactions mixing leads to less than optimal performance. This paper proposes some possible solutions to achieve better resource distribution, better control of transactions, performance and increased efficiency.

11:30 – 1:00 Lunch

1:00-2:00 "Performance Metric Validation in Chaos"
Bob Johnson, Landmark Systems Corporation

A decade ago, persons entering the performance and capacity planning disciplines could keep up with developments by reading publications, attending a few conferences, measuring their systems, and making decisions. Today, technology is growing more rapidly with each passing month. Hardware, software, and microcode change our business every day. Some even measure change in web years – defined as 1 month! We are constantly being asked to evaluate these changing systems without time to study and completely understand the system and businesses are making decisions based on these measurements which have a high probability of error. The first part of the paper describes the application of statistical classification theory to computer metrics. The application of statistical classification algorithms has helped many other scientific disciplines and can also be used to improve our understanding of computer metrics in real-world environments. The second part shows these techniques used in a bench marking setting to evaluate different configurations. The third part shows these techniques used in evaluation of a standard performance problem, such as: Do we need more central storage?

2:00-2:15 Break

2:15-3:15 "The Value of IT: A Strategic Perspective"
Rick Lebsack, IBM Corporation

This presentation will present an approach to quantifying IT Value and how it can be used to evaluate alternatives for IT investment. The first part is a tutorial/baseline for the relationship of Business Strategy to IT Strategy and how they relate to business value (The IT Challenge). This feeds into a Strategy Map, which documents the primary objectives toward which the client’s business is working. From this base, we expand into the measurement/quantification of value and why more than ROI is needed. With this level set, the second part is a hypothetical case study to evaluate a technology investment (Parallel Sysplex) and develop a business case using extrinsic and intrinsic components and how to fit these valuations into the individual client/customer’s evaluation model. This process raises the discussion of any investment alternative from a technical discussion to a business focus. While this does not alleviate the need for technical justification, it does broaden the parameters upon which the decision is being based. The beauty of this process is that it can be customized to each client with just a little research. From annual reports, web pages, or other sources, business objectives/strategies are defined. The IT strategies are mapped from conversations with directors or CIO’s. Within this framework any investment alternative or direction can be evaluated based on strategy.