8 Replies Latest reply: Mar 6, 2016 5:29 PM by Peter Cammaert RSS

    QV caching

    Ali Hijazi

      Hello

      working at the premises of a client

      the client has a server machine with 256 GB

      memory consumption is 90%

      this is because Qlik caches calculations and other stuff no?

      is there a configuration on the QMC via which I can change the consumption to be 60% instead?

      Please advise

        • Re: QV caching
          Clever Anjos

          Yes, please check your QMC

          System / Setup / [Select QlikView Servers] then Performance/ Working Set

          Capturar.PNG

          • Re: QV caching
            Andrew Whitfield

            As per the above, this is exactly how it's supposed to work. Is there an issue that you haven't mentioned?

             

            Andy

            • Re: QV caching
              Ashfaq Mohammed

              Hi,

               

              What version of Qlik You are on?

              Why not try by migrating to new version.

               

              For temporary fix

              If this is just Caching of ram. I would prefer to restart qlikview server service via batch file at specific interval to clear cache if any.

               

              Regards

              ASHFAQ

              • Re: QV caching
                Peter Cammaert

                If you leave the QMC->Working Set Limit->High value set at 90%, that means that your QVS is allowed to consume 90% of the total RAM amount in your server. 256 * 0.9 = about 25GByte is left untouched.

                 

                All other QlikView services including any occasional Reload job run outside of this memory allotment. Let's say that your OS needs about 2GB of RAM, then your QlikView services and reload processes (QVB.exe) should do their magic in about 23GB of RAM. Is that enough? If not, you should lower your Working Set Limits so that more RAM is set aside for other services and reload tasks.

                 

                BTW you can monitor RAM consumption either live using Task Manager, or programmed for extended periods using Performance Monitor or another Windows resource monitoring tool.

                 

                Peter