What is the best way to interpret Perfmon analysis into application specific observations/data?

梦想的初衷 提交于 2019-12-05 02:15:03

问题


Many of us have used Perfmon tool to do performance analysis. Especially with .Net counters, but there are so many variables going on in Perfmon, that it always becomes hard to interpret Perfmon results in to valuable feedback about my application. I want to use perfmon, (not a tool like Ants Profiler etc) but how do I accurately interpret the observations?

Any inputs are welcome.


回答1:


I use the Performance Analysis of Logs (PAL) tool: http://pal.codeplex.com/

It's not an "official" Microsoft tool, but I believe the author works for Microsoft. The project seems to be fairly active.

In addition to the canned threshold files provided (which are pretty good), you can write your own thresholds to analyze what your app needs. The generation of the HTML report with charts is also very nice.

UPDATE: PAL 2.3.2 no longer depends on the MS LogParser or MS Office Web Components; it uses PowerShell v2.0 or greater, MS .NET Framework 3.5 SP1, and the MS Chart Controls for .NET 3.5.



来源:https://stackoverflow.com/questions/37425/what-is-the-best-way-to-interpret-perfmon-analysis-into-application-specific-obs

易学教程内所有资源均来自网络或用户发布的内容,如有违反法律规定的内容欢迎反馈
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!