2 Replies Latest reply on Feb 17, 2011 2:53 PM by PHCharbonneau

    capacity planning - determine Weblogic memory based on available RAM

      can someone please let me know how to determine the amount of memory that can be used by weblogic server on a host that has say 32 gb ram,or 64 gb ram?
      is there a weblogic recommended calculation to figure out this number and later on tune the server based on application req?

      also, memory calculation based on processor bit size (32, 64 bit processor) and based on number of processors?

      thanks in advance
        • 1. Re: capacity planning - determine Weblogic memory based on available RAM
          Jay SenSharma MiddlewareMagic

          What should be the Heap size cannot be determined by the RAM size 32GB or 64GB. It depends on the Application...Application Functionality, How many Long living objects it creates , how many JNI calls it makes, How many Jars/Classes it loads, How many users requests your application at peak load, what kind of Caching you are using...etc

          Above things are the points which can be used to determine the required Heap Size and the JVM tunning options.

          There is no standard calculation or Formula available...The only option to determined the required Tunning parameters and JVM settings is...Load Testing and Performance Testing.

          Jay SenSharma
          http://middlewaremagic.com/weblogic/?page_id=2261  (Middleware Magic Is Here)
          • 2. Re: capacity planning - determine Weblogic memory based on available RAM
            Hi user8688761,

            Jay brought some very good points and I could not more agree. Proper capacity planning analysis is key in order to do that type of analysis.

            If you are interested in 32-bit vs 64-bit VM, our team recently did a comparaison analysis for one of our environment and you can still review the results and what to expect below but you really need to do your own analysis.


            You physical RAM requirement will also depend on the # of managed servers that you are planning to deploy.