May 20, 2009 at 12:13 am
Comments posted to this topic are about the item Method to the Madness
May 20, 2009 at 8:26 am
Wow...maybe the lack of responses supports your idea of needing a clear methodology for performance tuning. I'm still stuck in 2000 so my resources are limited. Profiler is usually my first tool. I know in the heat of battle, I tend to focus more on what tools to use instead of following a set method, which I guess ideally should be independent of version, or even environment for that matter.
May 20, 2009 at 9:18 am
I think part of the issue for XEvents and waitstats is that there isn't much documentation out there on them. MS came up with these features, didn't support them well, and only now do I see some authors bringing out information in the last month or two on them.
We could use a nice methodology for tracking down issues, though I think it would be a complex decision tree to go through. Maybe someone will take up the challenge.
May 20, 2009 at 9:39 am
"Instead of piling on more, often increasingly complex, tools, I believe what is needed is a clear roadmap and methodology for systematically addressing SQL Server Performance issues."
This is the most on-the-mark statement I have heard come from a SQL guru in a long long time. God bless you Tony for saying it. However, witnessing that Microsoft seems to love complexity over simplicity and often introduces tools and features that are never fully explained, I fear we are doomed to just more of the same. If there was any way to get the SQL community to settle on a good 'standard' (if possible) methodology and drive Microsoft and the user community that way, I'd be all for it.
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply