* [linux-lvm] walltime of pvs vs pvdisplay with 256 PVs
@ 2006-11-16 2:54 Mike Snitzer
0 siblings, 0 replies; only message in thread
From: Mike Snitzer @ 2006-11-16 2:54 UTC (permalink / raw)
To: linux-lvm
I have a testbed with 256 PVs. Running 'pvdisplay -c' takes .672 sec.
Running 'pvs' takes 43sec! I've tried this on RHEL4U3
(lvm2-2.02.01-1.3.RHEL4) and RHEL4U4 (lvm2-2.02.06-6.0.RHEL4). lvm2
from U4 is actually slower than U3 on this issue (1m vs 43s).
With RHEL4U3, pvs -vvvvv shows that a single device gets opened 264
times yet pvdisplay -c -vvvvv only opens a single device 8 times.
With RHEL4U4, there are 261 opens for pvs and 5 opens for pvdisplay
-c. Clearly there is something flawed in the pvs algorithm. I can
dig deeper but I wanted to put this info out there in case others
might have some insight.
please advise, thanks.
Mike
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2006-11-16 2:55 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-11-16 2:54 [linux-lvm] walltime of pvs vs pvdisplay with 256 PVs Mike Snitzer
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).