September 27, 2006 at 8:04 am
Hi All,
I have a new server which I finally installed, I finally got round to configuring the disks, see below:
| Database | Raid_Config | Drive |
| PersonDB | Raid5 | K |
| PersonDB_Log | Raid1 | J |
| Tempdb | Raid1 | J |
| Systemdb | Raid5 | K |
Drive K is quite huge, around 200GB and it sits on a SAN.
I noticed that queries now take ages on this new test server, I believe its something to do with the disks, what do you guys think.
Regards
T
September 28, 2006 at 3:45 am
John,
Personally I tend to go the same route as you for disk allocation, i.e. Logs on RAID1 (fault tolerance with normal read and write performance) and data on RAID5 (fault tolerance with fast read performance and slow write performance).
You say you have changed to a new server, what are the specs of the old one and the new one? Also are there any major differences in usage or the server (i.e. multiple SQL instances or even multiple uses or the server, i.e. being used as an application server as well)?
What size database are you playing with, have you rebuilt any required indexes, try running (DBCC SHOWCONTIG against the database) ? Are you querying the database directly or via an application?
I would suggest running Perfromance monitor on the server to ascertain some baseline performance results over time, run these for a good length of time, capturing queit and peak times and save the results. As a start you want to be capturing performance results for CPU (Processor) Usage, DISK I/O Usage and Memory Usage. This will help you rule out which subsystem is causing you issues.
Lets find this bottle neck.
ll
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply