increase of memory in sql server 2000 fro 2 to 6GB measure memory usage
HI,
I have increase the memory of my sql server 2000 SP3 from 2GB to 6GB due to high disk usage.
Can you explain why, even with my smaler memory i had a cache hit ratio of 100% but disk que lenth of more than 2000?
With adding more memory, the disk queu length has decreasedbut the % disk busy remains very high.
We have a IBM blade server with a disk attached to the EMC san
dababase size is:
1 of 69GB
1 of 69gb
1 of 19gb
1 of 8gb
How can i check if i have to much or to little memory?
/attached is a report of my memory and disk usage (The swap has been done on 02 july at 10 AM
20090703-4-.pdf
20090703-7-dbs05diskactivity.pdf
I have increase the memory of my sql server 2000 SP3 from 2GB to 6GB due to high disk usage.
Can you explain why, even with my smaler memory i had a cache hit ratio of 100% but disk que lenth of more than 2000?
With adding more memory, the disk queu length has decreasedbut the % disk busy remains very high.
We have a IBM blade server with a disk attached to the EMC san
dababase size is:
1 of 69GB
1 of 69gb
1 of 19gb
1 of 8gb
How can i check if i have to much or to little memory?
/attached is a report of my memory and disk usage (The swap has been done on 02 july at 10 AM
20090703-4-.pdf
20090703-7-dbs05diskactivity.pdf
Please indicate the size of the database on which queries are ran.
ASKER
Ok thanks just a few little questions to understand it better,
where exactly can i check the concurrent requests?
And why is it that the disk queue length is much less?
Why don't i see a difference to the cache hit ratio % (if it is above 90% doesn't it mean it runs totaly in the ram and not on disk?)
where exactly can i check the concurrent requests?
And why is it that the disk queue length is much less?
Why don't i see a difference to the cache hit ratio % (if it is above 90% doesn't it mean it runs totaly in the ram and not on disk?)
ASKER
It is a database of +- 96GB and 1 of +-69GB
Is it also normal that the SQL recompilations went up from +-100 pro minute to +-400 pro minute
Is it also normal that the SQL recompilations went up from +-100 pro minute to +-400 pro minute
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
That is because the increase memory has allowed the server to run more queries and therefore request more data from the disk. I'd say that your server works better because it actually processes more concurrent requests but this had a bad impact on IO contention. I'd suggest revising your file/filegroup strategy to get rid of that contention.