Performance monitoring - Windows server resources

I have been doing some research on what system monitoring is useful during performance testing.

I have concentrated on a minimum useful set of information needed to verify that no bottlenecks occur when using a performance testing tool. Even this minimal set should be useful though to enable a rough prediction of the maximum load that could be supported by the system under test.

There’s a wealth of information available on this subject. Two sites I found particularly useful were these:
http://www.sql-server-performance.com/tips_performance.asp
http://www.computerperformance.co.uk/HealthCheck/

One subject on which there seemed to be a range of opinions was whether it is best to set up monitoring remotely or locally. If you set up remotely there will be some network traffic, but if you set up locally there is the overhead of running perfmon itself. Do you think it is significant either way – if so, in what situations? Perhaps this is only an issue if you are already close to resource utilization limits.

One thing which is probably always true is that it’s best to be selective and only collect the performance counters you need. This document is my attempt to propose a subset which will nearly always be useful. Take a look and let me know what you think. Is there anything obviously missing? Is it too comprehensive?

Browse by topic:

Comments

I strongly believe there is no such thing as too much monitoring data. In some instances, setting up and running a large scale performance test is so large and costly that you only have maybe 2 or 3 kicks at the can. During these runs, you need to collect as much info as you can get.

One thing that I missed in your list was the process data, I find that particularly useful when I am hunting for "strange" behaviour.

You also might want to think about defining thresholds. MS has several documents that talk about reasonable thresholds for their counters.

Roland Stens
rstens@performancetester.com
http://www.performancetester.com

Roland - thanks for your thoughts and I will be on the lookout for the things you mentioned. If the note gets expanded to cover how to analyse these counters - and what thresholds to accept - I guess it will be just by adding more cross references to work others have already covered.

The document was geared to situations where repeated test runs of relatively short duration are envisaged. On reflection I did not make that very clear. I agree that if you are expecting limited testing, you had better not miss anything.

I will send out a note as and when the document gets updated. Given current commitments, this could be a while ....

David Cater

Acutest Ltd - UK Performance testing