[HPADM] Disk Bottleneck Question
From: Johnson, Craig E (Craig.E.Johnson_at_icn.siemens.com)
Date: 01/27/04
- Previous message: Naylor, Jim: "[HPADM] RE: -SUMMARY- Clarification on swapinfo/glance output"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
To: "'hpux-admin@dutchworks.nl'" <hpux-admin@DutchWorks.nl> Date: Tue, 27 Jan 2004 14:06:24 -0800
I have an RP5470 (L3000) with an external HP JBOD device connected through
a RAID 4Si card. The JBOD device is only half populated (one side, seven
disks), and it is configured as a RAID 5 LUN through "irm".
This is used as a Clearcase development server and most of VOBS are located
on this volume. When I look at the I/O by disk using Glance, I see that
when in use, this disk is active 100% of the sampling interval, and often
the request queue is quite high (over 50). The help pages from Glance
indicate that if the activity is greater than 50% for the interval, there
may be a bottleneck, however, I disagree, I think there's nothing wrong with
a busy disk. The problem I believe is the growing request queue.
The system is running the 64-bit version of HP-UX 11i. All filesystems are
VxFS.
Is there a way to tune that RAID 4si card? Can cache or memory be added to
it to improve performance? Anything else I should be looking at?
Filesystem tuning? Thanks!
Craig Johnson
Systems Administrator/Sr. Engineer
Siemens
--
---> Please post QUESTIONS and SUMMARIES only!! <---
To subscribe/unsubscribe to this list, contact majordomo@dutchworks.nl
Name: hpux-admin@dutchworks.nl Owner: owner-hpux-admin@dutchworks.nl
Archives: ftp.dutchworks.nl:/pub/digests/hpux-admin (FTP, browse only)
http://www.dutchworks.nl/htbin/hpsysadmin (Web, browse & search)
- Previous message: Naylor, Jim: "[HPADM] RE: -SUMMARY- Clarification on swapinfo/glance output"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|
