Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
tybaltzeke
Starting Member
28 Posts |
 Posted - 2010-03-07 : 22:45:44
|
| I am evaluating a TSQL script that updates a fact table by doing 14 separate queries. any help would be appreciated, and of course i realize the answers to my questions have to be general, very rough approximations because i'm not providing table sizes, etc. The entire process takes about 7 minutes to run (the tables aren't that big) but the client would like it to run in 2-3 minutes so that the cube can be refreshed in near real time during the trading day. how much performance improvement is possible by calling these scripts in parallel in execute sql tasks/execute package tasks? with one cpu? several cpus? how much performance improvement is possible by rewriting these scripts as data flow tasks (ole db inserts and updates) with cached lookups for the joins (the tables beling looked up are small enough to be cached effectively)?what kind of performance penalty might be encountered by the fact that aggregations perform worse in SSIS than TSQL? thanks to all in advance.. |
|
|
|
|
|
TSQL to SSIS, general performance considerations