AppliedOperations32 min
Performance and capacity management
Why work is slow and how much the machine can take. Response time and throughput, CPU, I/O and storage bottlenecks, MSU and the four-hour average, capping and software cost, capacity planning, batch window tuning and a method for investigating problems.
What you will be able to do
- Explain response time, throughput and how queueing makes busy resources slow
- Identify CPU, I/O, storage and contention bottlenecks from RMF and SMF evidence
- Describe MSU, the rolling four-hour average, capping and their link to software cost
- Run a structured performance investigation and avoid common anti-patterns
Lessons
- Response time, throughput and bottlenecksPerformance means two things: how fast one piece of work finishes (response time) and how much work gets done per unit of time (throughput).
- MSU, capping and capacity planningMainframe software cost is often tied to processor use measured in MSU, commonly the peak rolling four-hour average per LPAR.
- Investigating performance problemsGood performance work follows a method: define the problem precisely, compare with a baseline, find where the time goes, change one thing, and measure again.
Track your progress and earn a certificate.
Free account, 15 quiz questions for this subject.