IBM REDP-4285-00 User Manual
4285ch02.fm
Draft Document for Review May 4, 2007 11:35 am
70
Linux Performance and Tuning Guidelines
Figure 2-16 A sample Capacity Manager report
The Report Viewer window enables you to select the different performance counters that
were collected and correlate this data to a single system or to a selection of systems.
were collected and correlate this data to a single system or to a selection of systems.
Data acquired by Capacity Manager can be exported to an HTML or XML file to be displayed
on an intranet Web server or for future analysis.
on an intranet Web server or for future analysis.
2.4 Benchmark tools
In this section, we pick up some of major benchmark tools. To measure performance it’s wise
to use good benchmark tools. There are a lot of good tools available. Some of them have all
or some of the following capabilities
to use good benchmark tools. There are a lot of good tools available. Some of them have all
or some of the following capabilities
Load generation
Monitor performance
Monitor system utilization
Reporting
A benchmark is nothing more than a model for a specific workload that may or may not be
close to the workload that will finally run on a system. If a system boasts a good Linpack
score it might still not be the ideal file server. You should always remember that a benchmark
can not simulate the sometimes unpredictable reactions of an end-user. A benchmark will
also not tell you how a file server behaves once not only the user access their data but also
the backup starts up. Generally the following rules should be observed when performing a
benchmark on any system:
close to the workload that will finally run on a system. If a system boasts a good Linpack
score it might still not be the ideal file server. You should always remember that a benchmark
can not simulate the sometimes unpredictable reactions of an end-user. A benchmark will
also not tell you how a file server behaves once not only the user access their data but also
the backup starts up. Generally the following rules should be observed when performing a
benchmark on any system:
Use a benchmark for server workloads: Server systems boast very distinct characteristic
that make them very different from a typical desktop PC even though the IBM System x
that make them very different from a typical desktop PC even though the IBM System x