Forum Replies Created

Viewing 12 posts - 1 through 13 (of 13 total)

  • RE: C# CLR performance

    wait_typewaiting_tasks_countwait_time_msmax_wait_time_mssignal_wait_time_ms

    CLR_AUTO_EVENT 52285838414020656

    CLR_MANUAL_EVENT ...

  • RE: C# CLR performance

    It is usually done with 4 threads, but for test purpoases everything I've mentioned previously is done with only onw thread.

    Results from you qeury

    task_addressstatelast_wait_typestateforced_yield_count

    0x000000086E15C508SUSPENDEDCLR_JOINE_TASK_ATTACHED_TO_CLR0

    0x00000005801E0E08RUNNINGCLR_MANUAL_EVENTE_TASK_ATTACHED_TO_CLR0

  • RE: C# CLR performance

    C# CLR code: This is an excerpt since the actual code is probably 1000+ lines.

    An example of what is return from the LoadReports() and loaded in the ReportInfo dictionary

    #Id ...

  • RE: C# CLR performance

    -->HowardW

    Yep, I'll try to post the code in the next post.

    Btw. I tried running the code as a console app with the following results

    100.000 rows read to memory and the...

  • RE: C# CLR performance

    opc.three (4/19/2011)


    Something else worth checking immediately after an instance restart, as your process is running and after your process completes:

    -- credit to Bob B: http://social.msdn.microsoft.com/Forums/en-US/sqlnetfx/thread/bbc1c753-c5df-4d52-a567-c982cf506051

    SELECT SUM(pages_allocated_count) / 128.0 AS...

  • RE: C# CLR performance

    I'm always considering to re-consider 🙂 But I don't think I will rewrite it in TSQL, my thought was to make it a console application and run it with xp_cmdshell...

  • RE: C# CLR performance

    I tried setting MemToLeave in increments without any difference in execution time. I even tried setting it to 2GB without noticing any performance gains.

  • RE: C# CLR performance

    -> HowardW

    Yes, the SQL server is 64bit

    Doing the calculations in .Net does not remove parallelism since you can expicitly specify that in the code.

  • RE: C# CLR performance

    -> Incindium

    I tried running the code with smaller batch sizes with the following results

    [Num rows][local time][db server time]

    500 0.12s ...

  • RE: C# CLR performance

    I understand that but I don't like to leave things hanging without a reasonable explanation as to why this behaviour is encountered. It still seems strange to me to say...

  • RE: C# CLR performance

    Hi,

    I guess I probably could have used TSQL instead of the CLR in this scenario but I still think there is something lurking in background somewhere which makes the code...

  • RE: C# CLR performance

    Hi,

    I understand what you are saying and I would have thought that it could have been a data transfer problem. But the data is stored on the remote db server...

Viewing 12 posts - 1 through 13 (of 13 total)