on ‎2016 Jan 19 1:14 PM
Hello,
I have a couple of questions on MDXTEST
1. When I execute an mdx statement extracted from an excel report the first time it takes 45 seconds, if I execute again with the option to 'do not use cache' ticked it takes only a couple of seconds. If I exit the transaction and re-enter and try again then again it takes 45 seconds. What is the reason for this behaviour? Is there an option to stop it? The BPC Excel report itself when refreshed consistently takes about 45 seconds (we have turned off OLAPcache on the cube as it was causing performance issues) I have run an SQL trace using st05 and the first time it is querying the cubes, the second time it isn't.
2. When I analyse the SQL trace I can see that the SQL queries are running sequentially, is there any setting to change this behaviour to parralelize the sql queries generated from a single MDX statement?
Any help is much appreciated.
Many Thanks
Mike Tibbit
Request clarification before answering.
Hello,
The symptoms you have described seems like after the initial time the results are returned form some internal cache memory. To me that explains the difference between 45 second the initial time and 2 second the second time.
As for the second question, I think if there is a way to make MDX queries run in parallel it would have to be at lower level than BPC. For example, it might be possible at OLAP BAPI execution level. However, having said that I doubt that a single MDX query can be executed in parallel.
Best Regards,
Leila
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
| User | Count |
|---|---|
| 7 | |
| 6 | |
| 4 | |
| 2 | |
| 2 | |
| 2 | |
| 2 | |
| 1 | |
| 1 | |
| 1 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.