On Jan 4, 2010, at 10:35 AM, Thomas Burgess wrote:
slightly outside of my price range.
I'll either do without or wait till they drop in price....is there a
"second best" option or is this pretty much it?
If you need the separate log, then you can figure the relative latency
gain for latency-biased workloads by examining the devices in the
main pool and compare to the proposed separate log. If your main
pool is comprised of large, slow HDDs, then figure on the order of
10ms or so (or look at the datasheet and add rotational latency to
average seek). If your SSD is rated better, then it should be an
improvement.
Important note: your separate log does not need to be very large.
1 GByte can be more than enough to cover most cases (BobF's
workload excepted ;-) With the current b130 you can remove a
log device, so you can experiment with your workload on a small
slice and see if the log device helps. OTOH, if zilstat shows all
balls (zeros), then you can blissfully forget about separate logs.
-- richard
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
