Announcement

Collapse
No announcement yet.

Multiple Core Experience

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Carol... Your screenshot seemed to indicate a load spread more so than Tom's. Observing during market hours with a real streaming load would be interesting.Can you get QCharts at, let's say, 60% sustained on your x64 box?The same X%, then more %, then more% observations as detailed above would be interesting.The two sets of observations, a 32bit quad-core and a dual-core 64bit could be instructive.Again, I don't have a multi-core box to play with, to try this and that, so y'all please bear with my questions.Thanks,LAM 2/18/08
    Larry,
    Give these pics a look and see if it's telling you what you think it will? Note this from 6.0.1.16.

    Carol
    Attached Files
    Last edited by ckelly; 02-25-2008, 07:57 AM.

    Comment


    • #32
      Originally posted by ckelly
      Larry,
      Give these pics a look and see if it's telling you what you think it will? Note this from 6.0.1.16.
      Carol
      Hey Carol,

      Thanks much for taking the time to do the captures. Your 64bit sure seems to spread the load. I'm assuming that you were running your hochimema macdaddy wksp judging from the pegged and near pegged states.

      I had hoped to get to creating a series of workspaces last weekend. But alas, stuff came up. Ya know, a set of standardized documented workspaces and a test plan. At a designated time multiple folks could run the same workspaces according to a test plan that would increase load over time, both quotesheet and screen painting loads. Then compare notes and profile how different boxes tolerated the same loads by running the test scenario at the same time. This is not a trivial pursuit to get workspaces built, develop a plan, and get a group of testers together and on the same page regarding the plan and following the same documentation procedures.

      Thanks again, and to be continued.

      LAM

      Comment


      • #33
        Larry,
        Yes - yesterday's sample was with the macdaddy workspace. Something interesting occurred to me - it's not the size of the workspace that is the issue, I don't think. It's the time it takes QC6 to "recover."

        I ran the same large workspace again this morning with particular attention to "recovery" time. That is, let it "idle" (no mouse movement for several minutes) and take a couple of readings. Then, change symbols via the quote sheet and see where the CPU readings go - and then see how long it takes to "recover." This latest reading is of real interest to me. As you can see here, almost 2 minutes after the last mouse touch, QC6 is still trying to "recover." (I'm thinking this is much of what impacted yesterday's readings.) It almost seems that internally, QC6 is holding onto something instead of letting go and this causes the CPU to continue to churn unnecessarily.

        What do you think?
        Carol
        Attached Files

        Comment


        • #34
          Hi Carol,

          I have noticed the same effect but I haven't been able to measure it effectively. It feels like it is just taking a long time to recover after only just moving the mouse or selecting another stock on a big workspace.

          I have also noticed using 6.0.1.6 that during market hours, the Weekly and the Daily charts disappear although the intraday charts display correctly. (I will also post this in the 6.0.1.6 beta area unless it is already reported.)

          Thanks,
          Tom

          Comment


          • #35
            Originally posted by ckelly
            Larry,
            <snip>
            What do you think?
            Carol
            Hey CK,

            See comments in the 6.0.1.16 Beta feedback thread.

            LAM
            Sisyphus is a DayTrader

            Comment

            Working...
            X