Hi, David,
the report execution time can be vastly improved by having the WHERE clause tables arranged in order from the biggest table to the smallest one (as in number of rows). In some of our reports re-arranging tables reduced execution time from hours to seconds (literally). Good luck !
Cheers, Irena
Restrict the data coming back as much as possible with conditions. Do all aggregation on the server side instead of in the report (this can result in huge time savings!)
the
archives. If aggregating on the server side is still the best way…how do we accomplish that? Thanks
I recently posted about this issue. Walter said that he thought sorting the data BEFORE it is returned would improve computation speed. If you have a sort within your report on last name, sort by last name within the query. I tried it and it DID improve computation speed.
Basically, you would create aggregate objects (i.e., sum(Customer)) and use these in your queries. This way, you would return fewer rows than doing the aggregation on the client side (BusinessObjects) and BO would not have to do as much calculation.
Of course, if you had a data warehouse with prebuilt aggregations, you could just pull off the numbers you wanted without any calculations.
Restrict the data coming back as much as possible with conditions. Do all aggregation on the server side instead of in the report (this can result in huge time savings!)
the
archives. If aggregating on the server side is still the best way…how do we accomplish that? Thanks