Posted by Erland Sommarskog on 06/21/05 00:56

(hishamfangs@gmail.com) writes:
> I'm thinking this is an unusally long amount of time, and in a few
> months the number of records will qaudruple, so the problem will be
> even worse! This table was originally a few seperate tables and are
> accessed through joins, but I eliminated that since we wanted to boost
> performance and only inserted a minimalist amount of data into a single
> table just to return the above query, as expected, it boosted
> performance down to 1 minute-2 minutes, which was alot better than
> before, but I still believe that it can be better! I don't believe that
> it should take sql server 2 minutes to return 250,000 records grouped
> into 50! This query is ever only run once at a time, since there are
> only 5 users on the entire system.

As I understand your description, SQL Server will have to scan all
those 250000 rows to get the data.

But it would really help if you posted the CREATE TABLE and CREATE
INDEX statements for your table, as well as the query you use.

> Now, this table is indexed by date and by record id (primary key),
> and the server is running on a Compaq dual processor 950 mhz, with 2
> Gig of RAM, with 4 level 5 raid disks.

Hm, isn't that CPU a tad slow by today's standards?


--
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se

Books Online for SQL Server SP3 at
http://www.microsoft.com/sql/techinfo/productdoc/2000/books.asp

[Back to original message]