Guest
|
Posted:
Sat Oct 22, 2005 10:33 pm Post subject:
Re: LUN Size |
|
|
At work, I admin both IBM products: two ESS-800 ("Sharks") and
multiple FastT700 (DS4400) (in addition to HP and Network Appliance
systems) so I may be able to assist...
Here's the deal: With larger-and-larger controller cache sizes,
fibre-channel subsystems, and storage virtualization within modern disk
arrays, worrying about disk queues and more spindles is only for the
very specialized shops that wish to squeeze every single drop of
performance out of their block-level SANs; So that throws the old
religion of "more-spindles-equals-more-performance" out the window.
There's also no direct answer to your question, it's more about the I/O
behavior of your applications (Are there lots of random read/writes or
sequential reads? What kind of workloads are we talking about? Are
you aware of any hotspots on your SAN? Do you using RAID-5 or
RAID-10?)
Email or call me if you want to further discuss the issue...
Brian J. Mitchell
Systems Administrator
TRX
6 West Druid Hills Drive
Atlanta, GA 30329-2158 USA
http://www.trx.com
email: brian.mitchell@trx.com
office: +1 404 327 7238
mobile: +1 404 725 3128 |
|