Hello all --
I'm kind of new to this system administration thing and would appreciate
some advice on interpreting output from vmstat. We're running an Alpha
4100, 1 CPU, Digital Unix 3.2G. I think we have 512 Meg of RAM.
Here's a snapshot of the system under relatively heavy load for what we do:
Virtual Memory Statistics: (pagesize = 8192)
procs memory pages intr cpu
r w u act free wire fault cow zero react pin pout in sy cs us sy
id
4535 18 56K 367 6212 723 0 532 0 81 0 285 13K 901 61 35
4
4534 18 57K 95 6229 1108 30 597 0 283 0 244 13K 858 61 36
3
5534 17 57K 78 6226 928 0 528 0 272 0 315 13K 933 59 38
3
4535 17 57K 293 6192 576 0 526 0 38 0 247 12K 834 56 35
9
4536 17 57K 89 6200 703 30 576 0 39 0 100 14K 686 58 39
3
It seems to me that since the free list is relatively small, and paging is
occuring, some additional RAM would be in order with part of it configured
as cache. However, others in the office feel we should toss another CPU
in.
I'm not sure why there are so many threads waiting -- perhaps our NFS setup
is the culprit.
Anyway, I would appreciate any feedback on my interpretation.
Thanks,
Donn Aiken Applications Engineer Regents College
daiken_at_regents.edu
http://www.regents.edu
These comments are my own and do not necessarily reflect the
policies or position of Regents College.
Received on Tue May 05 1998 - 15:00:01 NZST