Bulk insert slowing select in different DB?
From: Danny (istdrs_at_flash.net)
Date: 10/28/04
- Next message: Shabam: "Too many connections"
- Previous message: Danny: "Table schema for DBCC show_statistics"
- Next in thread: Danny: "Re: Bulk insert slowing select in different DB?"
- Maybe reply: Danny: "Re: Bulk insert slowing select in different DB?"
- Messages sorted by: [ date ] [ thread ]
Date: Thu, 28 Oct 2004 01:21:30 GMT
We have a SQL 2K standard sp3a box with 4 procs, 4GB ram, and IO on a good
SAN. We had a situation where there were only two processes running on
server. Process one is a bulk insert into DB1. Process two is a simple
select in DB2 with join of 2000 rows with 3.5 million rows. The select
query plan is scan of the small table and a nested loop to an index seek on
the large table. Process two runs takes minutes where without the bulk
insert running takes seconds. Same query plan. CPU at 25%. Avg. Disk
Queue at fluctuating but averaging about 17.
Any thoughts on how a bulk insert into a separate DB on the same server
could cause a 20 fold decrease in query performance?
Thanks,
Danny
- Next message: Shabam: "Too many connections"
- Previous message: Danny: "Table schema for DBCC show_statistics"
- Next in thread: Danny: "Re: Bulk insert slowing select in different DB?"
- Maybe reply: Danny: "Re: Bulk insert slowing select in different DB?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|