SCMG Meeting Richmond
May 19, 2004

Location:

Agenda:

Speakers:

Chris Romano
Chris Romano as been with VMware for about 3 and half years. Prior to Joining VMware in November 2000 he was at the NASDAQ Stock Market/ NASD for 6 years in Corporate IT. Chris managed 70 Intel servers and 1000 desktops and in the early 90s was at a Novell Centric Systems Integration Company in Northern Virginia.

Tom Thompson
As Principal Technologist for OPNET Technologies, Tom is responsible for providing overall end-to-end performance and capacity solutions for enterprise customers, with emphasis on mainframe and server capacity planning. He is an IBM veteran with more than three decades of delivering performance and capacity solutions to major IBM customers. As a consultant in the IBM Global Services SNAP/SHOT organization he leveraged leading-edge methodologies, modeling, and analysis techniques to meet demanding customer requirements. Today he leverages a broad range of experiences to continually enhance OPNET solutions for enterprise customers.

Ivan Gelb
Ivan Gelb is President of Gelb Information Systems Corporation (GIS), a consulting firm that provides management and technical consulting services in the U.S. and internationally. His extensive Information Technology (IT) background includes: determination of optimum hardware and software requirements for mainframe and client-server systems; effectiveness evaluation of computer systems and related organizations; data communications systems design and implementation; computer systems end-to-end availability management, performance management and capacity planning; development of software packages and proprietary measurement data analysis techniques.

During his more than 30 years of experience, Mr. Gelb performed technical and management services for more than 100 organizations such as JP Morgan, Merrill Lynch, PepsiCo, FBI, State of California, New Jersey State Office of Information Technology, and New York City Board of Education. He is a speaker at various technical conferences, Director of Computer Measurement Group, article writer and editor for a number of trade publications.

James Cook
As Senior Software Consultant for BMC Software, James works primarily with the performance and capacity planning solutions for enterprise customers, with emphasis on distributed systems. Over the past 20 years he has been a Systems programmer, Unix administrator, PATROL administrator, performance analyst and capacity planner.

Linwood Merritt
Lin started his data processing career in 1970 as a Simulation Analyst and has worked at Capital One Services, Inc. in Richmond, Virginia since 2001. He has been a US CMG member since CMG84, has been the Project Manager of the "Enterprise Wide Capacity and Performance" project of SHARE since 1994, served as Subject Area Chair for CMG2003, and is serving as Assistant Program Chair for CMG2004.

Lin has published 14 CMG papers (not counting 1 additional for UKCMG 2003) and presented at 25 SHARE conferences. He won the J. William Mullen Award at CMG97 for the presentation "Performance Data from the Server to the Intranet: Getting the Data and Reporting It," and presented it at UKCMG in 1998.

Igor Trubin
Igor started his career in 1979 as an IBM 360/370 system engineer. In 1986 he got his Ph.D. in Robotics at St. Petersburg Technical University (Russia) and then worked as a professor teaching CAD/CAM, Robotics and Computer Science for about 12 years at the same University. Igor has published about 30 papers and made several presentations for different international conferences related to the Robotics, Artificial Intelligent and Computer fields.

In 1999 he moved to the US and since that time has been working at Capital One Services, Inc. in Richmond as a Capacity Planner. The first CMG paper was written and presented by him in 2001 in Reno, NV. The next one, "Global and Application Level Exception Detection System Based on MASF Technique," won a Best Paper award at US CMG 2002 and was presented again at UKCMG 2003 in Oxford, England.

Abstracts:

VMWare - What, Why, How?
Presenter: Chris Romano, VMware
VMware is a sofware solution for running mulitple OS images on a single set of hardware. This presentation will present the VMware technology and components. Topics to be addressed include what applications are good fits for this technology and what to expect in terms of performance and capacity when converting from standalone severs to VMware images on shared hardware.

Optimize Server Consolidations with End-to-End Modeling
Presenter: Tom Thompson, OPNET Technologies, Inc.

Consolidation or relocation of servers can have significant impact on end-user response times and on the network infrastructure. The increased costs of the network may significantly reduce the ROI that was expected. End-users may not experience satisfactory application response times. In this session we will present best-practice methodologies that meet these challenges with critical analysis and current modeling techniques to accurately size the required changes to the enterprise infrastructure - leveraging server configurations and workloads, network configurations and workloads, and network application characteristics. These methodologies can be used to proactively:

• Ensure adequate server capacity
• Ensure adequate network capacity
• Ensure end-to-end application performance
  

By following these methodologies early in the planning cycle, the user can more accurately calculate the costs and risks associated with the consolidation or relocation initiative - before significant expenses are incurred.
Presentation

CICS Performance Management 2003
Presenter: Ivan Gelb, Gelb Information Systems Corporation
Performance management controls of CICS Transaction Server (TS) for OS/390 – z/OS greatly affect performance and the effective capacity of a complex is impacted. This presentation focuses on the best practices for CICS controls and z/OS environmental factors which affect a CICS region’s overall performance, total required processor capacity, real and virtual storage, and disk input/output service. Workload Manager (WLM) definitions that may help or hinder CICS will be included. The potential risks and benefits associated with the selection of actual and default values will be identified. Samples of reports for health monitoring and problem analysis will be presented.
Presentation

Best Practices in Performance Assurance
No abstract available.
Presentation

A Finger in the Wind: Forecasting Techniques for Capacity Planning
Presenter: Linwood Merritt, Capital One Services, Inc.

Robust capacity planning entails the analysis of data to project future demand. This analysis can be as simple as a linear trend of historical demand data, and can be as complex as multivariate regression using business drivers combined with analysis of changing business functionality. This discussion explores the use of Excel and SAS to perform simple trend analyses of computer resources and produce capacity planning forecasts.

Disk Subsystems Capacity Management, Based on Business Drivers, I/O performance metrics and MASF
Presenter: Igor Trubin, Ph.D., Linwood Merritt, Capital One Services, Inc.

One site's experience of using business drivers and I/O performance data from SAS/ITSV and BMC performance databases to produce web-based disk subsystem capacity usage reports for a large, multi-platform environment.

The home made system captures global and application level I/O performance data and automatically publishes on an ITRANET web site the following information:

• Hourly disk I/O rate with trends and 6 month forecast charts correlated with business drivers forecast.
• Disk I/O channels capacity estimation as a threshold on the I/O rate charts.
• SPC charts based on MASF technique to show global and application level statistical exceptions of I/O resource usage.
• Tabulate data of the DISK I/O subsystem performance status colored automatically to show the possible performance or capacity issues regarding disk subsystems.

Presentation

Sponsors:

Enterprise companies face the requirement to align IT resources with the needs of the business in order to ensure the successful performance of business critical applications. Not only are IT organizations faced with meeting the performance requirements of these applications, they must also optimize the utilization and cost of the IT infrastructure. The HyPerformix solution helps reach this alignment; providing the decision-support tools to understand end-to-end application performance and predict the effect of changes BEFORE they are made. The benefits of the HyPerformix solution include: reduced risk of deploying enterprise applications, avoidance of unnecessary IT spending, optimization of IT infrastructure, revenue protection, and improved time-to-market. HyPerformix is a leading provider of performance and capacity planning solutions for the enterprise and is headquartered in Austin, Texas.

BMC Software enables you to monitor and manage business services, along with the technology infrastructure supporting them—applications, data, systems and networks—simultaneously.