From: Butler, Mathew (Mathew.Butler@logicacmg.com)
Date: Sun Mar 09 2003 - 19:56:29 EST
I am looking for some additional information on the Tru64 vmstat command.
At the moment I am seeing this kind of output from /usr/bin/vmstat -w 300
(on a 4 CPU AS4100 (T64 V5.1) with 4Gb memory)
Virtual Memory Statistics: (pagesize = 8192)
procs memory pages intr cpu
r w u act free wire fault cow pin pout in sy cs us sy id iowait
7 287 41 469K 1060 42K 2088 419 690 0 475 8K 5K 63 14 6 17
13 279 41 469K 1058 42K 1732 330 600 0 474 8K 5K 65 13 5 16
7 283 43 469K 1063 42K 1811 331 608 0 471 9K 5K 61 14 6 18
8 284 41 469K 1058 42K 2040 420 690 0 475 9K 5K 64 14 6 16
10 281 42 469K 1063 42K 6717 655 1149 0 482 11K 5K 70 15 4 12
13 280 40 469K 1064 42K 2109 320 590 0 482 10K 5K 73 14 3 10
I see some bursts of I/O on my swap device after heavy paging in.
I am looking to find out how I can
a) identify conclusively whether the server is CPU bound.
b) identify conclusively whether the server is Memory bound.
c) identify conclusively whether the server is I/O bound.
What does the iowait column signify and how do I interpret it?
I often see the runqueue with a value much greater than the number of CPU,
when shoul dthis bne considered a problem?
For a) - c) above which of the outputs from vmstat are most significant?
If somebody can reference some good doco on this is would be great (I've
tried the Performance guide and man pages, but am looking for something
else).
Mathew Butler
Consultant
LogicaCMG
17-19 Orion Road, Lane Cove West,
Sydney, NSW 2066
Australia
E: butlergm@logica.com
T: +61 (0) 2 9424 4623 (direct)
F: +61 (0) 2 9427 9490
http://www.logicacmg.com
This e-mail and any attachment is for authorised use by the intended recipient(s) only. It may contain proprietary material, confidential information and/or be subject to legal privilege. It should not be copied, disclosed to, retained or used by, any other party. If you are not an intended recipient then please promptly delete this e-mail and any attachment and all copies and inform the sender. Thank you.
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:10 EDT