1 Reply Latest reply: Dec 18, 2015 6:03 AM by Onno van Knotsenburg RSS

    What is the Role of RAM in Qlik....!!!

      Hi All,

       

      I have  a Few basic knowledge on Qlik , I want to know the Role of RAM in Detail with respect to Qilk. And in addition to this I would Like to ask a clarification that I have used qliksense with some Extension ( Tree Type View) for a hierarchical representation of data of size 28GB in a server where there is 128 GB RAM is available.

       

      Currently I have closed all the application after reload that I made last night , but when open the application , the performance is too low and the RAM is utilized to the Maximum.

       

      I have also Closed all the application related to qlik, but still the RAM utilization by QLik is to the fullest. Can any one let me know the reason for the RAM behaviour and if the data stores in the RAM how long (time) will it take to clear the infromation from RAM.

       

      What is Immediate Solution to Reduce the RAM utilization other that restating of services.

       

      Thanks In Advanvce,

      SATHIS

        • Re: What is the Role of RAM in Qlik....!!!
          Onno van Knotsenburg

          Qlik caches all calculations made up to a set maximum percentage of your RAM.

          Even if you close all application and wait till the end of time, this caching will stay in your RAM. This is perfectly normal and the way that Qlik is designed. This caching will result in faster performance when someone starts asking the same questions again.

           

          This should never be a problem, as you either

          A) install Qlik software on a dedicated server

          or

          B) adjusted the maximum percentage of RAM it is allowed to eat up.

           

          If RAM consumption is about 70% for your 128gb RAM machine, there is something to solve. Either your data model could use some improvements or the amount of data you load in your application is WAY to much for the memory you have (but we would be talking about billions of rows / many, many columns).