Tab navigation
- About this workshop- selected tab,
- How to enroll
Objectives
System performance is becoming increasingly difficult to manage. Many computer installations are experiencing problems with slow response times or poor throughput and the inefficient use of resources. Often the problems are the result of greater technical complexity, the growth of applications, and the challenge of having staff with the appropriate skills. To address these problems, the IBM ITSO is offering this performance immersion class. Using theoretical lectures and real life performance report analysis, the students:
· Learn key metrics for working with performance management
· Develop a deeper understanding of mainframe system behavior and integration to help tune the system with confidence.
· Improve their ability to plan and manage the resource growth of an installation to handle performance issues.
Description
During this class, we discuss a set of real performance reports from an anonymous customer. In addition, attendees can bring their own reports from RMF, Omegamon, or another monitor to be analyzed individually after class hours. The class discusses features available up through z/OS V1.13. The major topics covered in the class are:
· Processor performance, including Parallel Sysplex performance
· I/O considerations
· Central and virtual storage considerations
· Workload analysis
This class is up-to-date. Each new function in hardware (in a z196, for example) or in software (in z/OS V1R13, for example) affecting system performance is included in the class, with real life examples, if possible. Here are some recent topics:
· WLM Capacity Provisioning
· 1 MB address pages
· HiperDispatch
· MIDAW
· New FICON channel protocol (no more CCWs?)
· Performance recommendations on CF duplexing
· Saving MIPS
· Fact: z/OS is not now a fully preemptible operating system. What are the consequences?
· New discoveries about MSU/hr, MIPS, cycle time, ITR - do not miss
· How to manage the WLM goals of CICS, IMS/DC and WebSphere workloads.
· The enclave concept
· Trading central storage by CPU (such as DB2 buffer pool page fixing) and by I/O (such as VSAM SMB)
· How to encrypt without spending many CPU cycles
· Why we need zAAPs and zIIPs and how they are managed by z/OS
· What is a CMR delay during an I/O operation?
· Capping or not? If yes, what type?
· The consequences of the WLM aging algorithm
· How to improve DASD performance today
· The mysterious QDIO algorithm
· The possible revival of IRD dynamic channel management (DCME) for FICON channels
Target Audience
Clients, business partners, and IBMers are all welcome. The course is written for multiple companies to attend together.