3 Replies Latest reply on Dec 7, 2012 5:47 AM by Vasavya Chowdary

    calc script time is differing

      Hi All,

      i have calculation script normally it takes 10min time to execute.
      same calculation is taking now 20 min time.
      can any one help me in know y it is happening like that and what measure do i need to take for reducing time.

      thanks in advance.

        • 1. Re: calc script time is differing
          Couple of things that spring to mind

          1) Make sure that there were no outline changes that could have affected performance. Increased number of members, dynamic calc changed to stored, even someone messing with some server or db settings.
          2) If you calc script has multiple steps, look at each one in the log, compare the times between the 10 min run and the 20 min run. See if they all increased, or if only one step increased.
          3) Check cube fragmentation
          4) Maybe there were other things running on the server at the time the calc was run (exports, or other loads/calcs) and perhaps it was a "one off"?

          See what you can find out.
          • 2. Re: calc script time is differing
            totally agree, add two points: the database size become bigger than last time, or set some members into sharemember will be impact it too.
            • 3. Re: calc script time is differing
              Vasavya Chowdary
              for optimize: if ASo export and clear database and reload ,same with BSo but restusture and calc :(
              Enable tracking will help some sort

              even loading pattern : make it row format ..dense at last (BSo) ASo take highest memebrs in dimension list ....manythings come across
              check lLEVEL zERO DATA CHECK BLOCK SIZE

              Some times intermedite levels do get aggrigated , lets be frank .. usualy data is looked over level zero may be but top level usualy so if it not required to AGG to member in between dimension

              FYI data value stored only once, but it displays in multiple locations. Storing the data value only once
              offers considerable space saving as well as processing efficiency so FYI about shared member