Skip to main content
Software Engineering

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

23
  • 7
    Have you tried running an execution plan against your queries? Commented Oct 24 at 14:24
  • 11
    "The CPU usage stays very low (less than 5%)" - how many cores does the app server have? It might be a case of a single-threaded application using only one of them. Report generation sounds like something that should be trivial to parallelise. Commented Oct 24 at 22:47
  • 3
    You're talking about Microsoft, where 100% CPU utilization means all cores active. (Linux load average and other monitoring tools normally report 100% as one core busy, 800% as 8 cores busy, etc. Much more sensible especially for workloads that can't always use all cores.) When you say SSD 80% busy, are you talking about IOPS, bandwidth, or what? Or like 80% of the time there's at least one outstanding request? I assume this number is from Windows resource monitor or something; IDK what metric it uses. Commented Oct 25 at 3:24
  • 3
    After you optimized your tasks, I guess it would be an interesting read how you finally managed it. Commented Oct 25 at 5:07
  • 1
    @PeterCordes: relevant questions, indeed. While for the app server, I have RDP access, for the database one, the performances are collected by a special app, and it is unclear, what exactly does it display. This applies both to CPU (i.e. whether 100% would be for one core or all cores), and to the SSD (i.e. what exactly is being measured and what busy means in this context). Commented Oct 25 at 7:38