Re: Disk Performance issues?

The only way to get more I/O per user is to either add more disks to the LUN (more spindles= more I/O)
or reduce the I/O footprint of the existing users. MS recommends no more than 2500 items in the default
folders so that is contributing to your load. It sounds like you have all your databases on the single RAID 10
LUN... if so that is most likely creating alot of contention. We generally have a separate LUN for each SG
and limit each SG to two databases. How many SG's and databases do you have? Are you users using
alot of outlook addins or Blackberry devices?

Todd


"Brian Hoyt" <hoyty@xxxxxxxxx> wrote in message news:udhdzHt8GHA.4224@xxxxxxxxxxxxxxxxxxxxxxx
I ran ExPTA to help with a problem I was having. Some users get the pop-up that indicates that outlook is waiting for data from the exchange server. The output from ExPTA was that I have a disk bottleneck which was what I suspected. My problem is I don't know how to speed it up anymore? The server specs:
Dell PowerEdge 2850
6 GB RAM (even with PAE only 4 GB accessible?)
2xXeon 3 GHz HT (4 logical Proc)
Dell PERC 4e/Di
2x36 GB U320 15K Mirror - OS, Ex Sys, Logs
4x73 GB U320 15K RAID10 - DB only

According to the report I had a peak of 241 ms latency on the DB drive. During the monitoring I opened mailboxes that experience the waiting pop-up. These mailboxes have folders with 5K-20K items which I understand is quite large. I am just not sure if there is anything I can do to speed up disk access at this point? Any suggestions?



.

Relevant Pages

  • Re: Disk Performance issues?
    ... All our DB are on a single partition that lives on the single virtual disk ... made up of the RAID 10. ... LUN ... your databases on the single RAID 10 ...
    (microsoft.public.exchange.admin)
  • Re: Caching control
    ... |> | invalidate/unmap them in order to discard the data from memory. ... |> writing out to disk. ... | easy to discard as clean disk cache. ... stating that a specific amount of RAM can be used only for I/O ...
    (comp.os.linux.development.system)
  • Re: Dynamic configure max_cstate
    ... fio is a disk I/O workload ... which doesn't spend much time with cpu, ... I also thought it's related to timer. ...
    (Linux-Kernel)
  • Re: Dynamic configure max_cstate
    ... fio is a disk I/O workload ... I also thought it's related to timer. ... But oprofile data shows acpi_pm has more cpu utilization. ...
    (Linux-Kernel)
  • Re: Dynamic configure max_cstate
    ... fio is a disk I/O workload ... are we actually taking ACPI Cx exit latency into account, ... I also thought it's related to timer. ...
    (Linux-Kernel)
Loading