Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations IamaSherpa on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

MSTR & SQL Server 2000

Status
Not open for further replies.

NOSMAN

Instructor
Sep 8, 1999
3
ES
Hi Everybody, Currently we are testing a solution based on Windows 2000 / SQL Server 2000 and MS 7.2.3, We have a partitioned view( UNION ALL ) acting as a FACT Table composed about 30 tables with aproximately 30 Gb in amount, We've got 5 or 6 customized reports, and we note that the performance on any drill-down operation becomes unnaceptable ( Very slow ), and drives to timed-out requests,etc,etc.
Our solution is based on 2-Xeon CPU Servers at 3.4 Mghz with 3 and 4 GB RAM and working over a NAS Storage Solution.

There are any planning or start-up guide to fix or adjust the needed parameters to start-up rightly??
Is a good solution the platform Windows / SQL Server, or is a good advice to move an Oracle / Unix platform ?

Thanks in advance
NOSMAN.
 
suggest the following.

- instead of unionall with 30 tables, create a view for each table, and use those views in the MSTR project. The unionall might join unnecessary tables for your reports.
- take 2 or 3 drill down reports, look at the SQL and see why the performance is slow. See if it is just as slow if you run it through another querytool with the same sql.
- add indexes. In the SQL, look for the where clauses, see what is being used to filter the data. Make sure those columns are indexed.

that's it. There should be no problems with 30GB of data with MicroStrategy. The latest OLAP survey puts their median customer database size at 300GB!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top