Library
Performance Management
Considerations for data center efficiency based upon alternative Sysplex configurations - Hollander
AS LARGE SYSTEM COMPUTING ENVIRONMENTS INCREASE IN SIZE IT BECOMES MORE ESSENTIAL TO FOCUS ON A DESIGN THAT MAXIMIZES AVAILABILITY AND TRANSACTION THROUGHPUT WHILE MINIMIZING OPERATIONAL COSTS.
THE RIGHT SYSPLEX BASED CONFIGURATION CAN ALLOW MAINFRAME USERS TO ACHIEVE THEIR GOALS FOR LARGE SYSTEM COMPUTING. THIS PAPER DESCRIBES THE BASIC CONFIGURATION VARIANTS, THEIR BENEFITS AND HIGHLIGHTS SOME OF THE CRITICAL CONSIDERATIONS.
SELECTING AND REFINING THE OVERALL CONFIGURATION CONTRIBUTES TO THE SUCCESS OF THE STRATEGIC BUSINESS NEEDS BEING FACILITATED IN THE SYSPLEX ENVIRONMENT. EXPANDING THE SYSPLEX TO INCLUDE NEW WORKLOADS IN DISTRIBUTED SYSTEMS HELPS EXTEND THE RETURN ON INVESTMENT (ROI) OF THE SYSTEM Z HARDWARE. INCLUDING THE CONSIDERATIONS DISCUSSED HERE WILL ADD TO THE SUCCESS OF THE SYSPLEX.
PERCEPTION IS REALITY - The Psychology of Performance Management - Denise P. Kalm
Performance management is the most thankless job. If performance is good, the perception is that you are not needed. If response time is lagging, you are a failure. Even when the measurements look good, if the end user is unhappy, again, you have failed to meet your goals. Managing expectations, end users, colleagues, business managers and your boss – these are your real SLOs (service level objectives). Learn how to use ‘social styles’ and other techniques to manage change, make recommendations, handle hidden service levels and deal with degradation with greater success, to enhance your reputation, visibility and value to your company.
THE STORK CORRELATION - USE & ABUSE OF STATISTICS IN PERFORMANCE & CAPACITY PLANNING - Kalm
Terms like “mean,” “standard deviation” and “correlation” are used frequently in our industry, but how many really understand them? And how often are statistics misused, creating confusion and misunderstandings, when their main purpose is to communicate your results more accurately? This paper will include a review of pertinent statistics, how they are derived, and where and when to use them in performance and capacity planning.
Statistics can be fun and increase your decision-making confidence. Now, if we can just get past the terminology and the math…and remember, it is often more art than science.
“Statistics is a systematic method for getting the wrong conclusion with 95% confidence.”
JUMP START YOUR PERFORMANCE CAREER USING ANALYTIC MODELING - Denise P. Kalm
What is the career path for a performance analyst? A capacity planner? Most technicians paid their dues in the trenches of systems programming or operations. Having done it all, there can appear to be few interesting challenges left. Move up in your industry to become a strategic planner or a systems architect. Analytic modeling is the number one tool to turn a bit-twiddling performance geek into an advisor to the CIO. This paper will discuss the methodology to make modeling pay off in your career.
End User Performance Tuning - Denise P. Kalm
Most performance tuning activities involve systems tuning – managing I/O, adjusting parameters, moving workloads, etc. Only rarely, do you have the opportunity to work directly with the application developers to improve the code efficiency. But there is a third tuning method, one that is only rarely employed, but in many cases, can be very effective. This is particularly true when you can’t touch the application.
End user tuning – working with the user to improve their utilization of a specific application or business function – can result in more balanced system resource consumption, as well as more consistent response time. In this paper, examples and approaches to working with end users will be discussed.
WLM- After Goal Mode, the Yellow Brick Road Didn’t Lead to Emerald City - Norm Hollander
Early piece highlighting the challenges of moving to goal mode. A useful read for anyone, even if you think you got everything right. (Feb 2003)
How’s That Work? A Focus on WLM Goal Mode Batch Workloads - Norm Hollander
This session discusses various aspects of Batch Workloads in Workload Manager Goal Mode. You will see how these Workloads originate, guidelines for setting up, options for qualifying them, and best practices for defining the associated Service Classes.
(March 2005)
How’s That Work? A Focus on WLM Goal Mode STC Workloads - Norm Hollander
This session discusses various aspects of Started Task Workloads in Workload Manager Goal Mode. You will see how these Workloads originate, guidelines for setting up, options for qualifying them, and best practices for defining the associated Service Classes.
(March 2005)
Discovering Mainframe Performance Opportunities:When the World Is Not Spinning Fast Enough
Norm Hollander presents the following topics:
- Mainframe Environments continue to grow and become more complex at a rapid pace
- z/Series Hardware and Associated Peripherals are challenging for most Customers
- Keeping Up with z/OS (z/VM, z/VSE) Technology is Time Consuming
- The number of IT professionals at most Customer Sites has NOT grown with the Environment
- Many Customer Sites have discovered that their previously well-behaved Systems are just not performing well
- Adding New System Images may cause the Entire Environment to NOT perform well
- Creating D/R, Development, Q/A, Test, and Other Sandbox Environments may cause the Entire Environment to NOT perform well
- Legacy Applications may no longer be performing well
- Even installed products may be suffering
(April 2005)
How’s That Work? A Focus on WLM Goal Mode Transaction & Database Management Workloads - Holla
This section discusses various aspects of Transaction Workloads in Workload Manager Goals Mode. You will see how these Workloads originate, guidelines for setting up, options for qualifying them, and best practices for defining the associated Service Classes. (March 2006)
New Age Autonomic Computing - Linking Performance Monitors and Automation Solutions
In many of today’s Complex Data Center Environments, Customers are refining their Workload Manager Service Policies, Implementing Intelligent Resource Director (IRD), and investigating Enterprise WLM (EWLM), to help enhance their Continuous Availability and High Performance Environments. But, utilizing just these tools may not be enough to achieve the desired results. This session looks at some Autonomic Processes to link your currently installed Performance Monitors and Automation Solutions to increase the effectiveness for Self-Monitoring, Self-Configuring, and Self-Healing of Critical Business Applications.
(August 2007)
The z/Approach for Setting Up a z/OS Operating Environment - Norm Hollander
This presentation discusses a process for implementing z/OS into several Operating Environments. The Target Operating Systems are z/OS 1.4 and above. The goal is to create an environment to implement system software and products in a timely manner (potentially eliminating a need for IPL windows), to create an environment to implement system software and products using a systems management approach (Roll-out/Delivery Process) and to create an environment that facilitates exercising D/R processes. (May 2005)
New Age Performance Monitoring:Concepts and Directions- Norm Hollander
In today’s complex Data Center Environments, monitoring Performance is still an important function. How we performed this function in the past, was good for those old environments. Today, we must adopt new Concepts and Directions in order to provide effective Resource Utilization, and take the necessary corrective actions to protect our precious resources. This session will look at some of these New Age Concepts and Directions. (November 2008)
The Minimum Daily Adult - The Right Metrics and The Wrong Metrics - Denise P. Kalm
In capacity planning and performance analysis, we are inundated with metrics that purport to measure performance, but how we display them and understand them is what matters. Many metrics we take for granted are actually not that useful, and yet, permeate our world. This paper is intended to help you understand why the same-old, same-old metrics aren’t good enough, and what works better.
“If you want to inspire confidence, give plenty of statistics. It does not matter that they should be accurate, or even intelligible, as long as there is enough of them. “
~ Lewis Carroll
BROWSE LIBRARY
![]()
Application Management (0)
![]()
Capacity Management (2)
![]()
Database Management (0)
![]()
Databases (0)
![]()
General (11)
![]()
Network Management (1)
![]()
Performance Management (14)
![]()
Security (0)
![]()
Storage (1)
![]()
Systems Management (12)
![]()
Workload Automation (0)
![]()
