March 3, 2004 at 4:54 am
Hi!
I've created a separate mirror for indexes of my databases. It has very large disk queue lenth
for small data transfers (when writing data):
Avg. Disk Queue Lenth 20-100
Disk Bytes/sec less than 1MB
I have another mirror, that is on exactly the same type of disks, it holds tempdb and all logs of all
databases. It performs much better:
Avg. Disk Queue Lenth 3-5
Disk Bytes/sec 10-15MB
Write cache is enabled on both volumes.
What is wrong with my indexes volume?
Thanks.
March 3, 2004 at 7:15 pm
disk compression is one place to look at.
Thinking disk partition cluster size may also play a role.
March 4, 2004 at 2:31 am
There is no disk compression. Cluster size is 64k on both mirrors.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply