From: | Chris Browne <cbbrowne(at)acm(dot)org> |
---|---|
To: | pgsql-performance(at)postgresql(dot)org |
Subject: | Re: Slow count(*) again... |
Date: | 2010-10-12 16:03:57 |
Message-ID: | 87iq179z4i.fsf@cbbrowne.afilias-int.info |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers pgsql-performance |
sgendler(at)ideasculptor(dot)com (Samuel Gendler) writes:
> Geez. I wish someone would have written something quite so bold as
> 'xfs is always faster than ext3' in the standard tuning docs. I
> couldn't find anything that made a strong filesystem
> recommendation. How does xfs compare to ext4? I wound up on ext4 on
> a dell perc6 raid card when an unexpected hardware failure on a
> production system caused my test system to get thrown into production
> before I could do any serious testing of xfs. If there is a strong
> consensus that xfs is simply better, I could afford the downtime to
> switch.
It's news to me (in this thread!) that XFS is actually "getting some
developer love," which is a pretty crucial factor to considering it
relevant.
XFS was an SGI creation, and, with:
a) the not-scintillating performance of the company,
b) the lack of a lot of visible work going into the filesystem,
c) the paucity of support by Linux vendors (for a long time, if you
told RHAT you were having problems, and were using XFS, the next
step would be to park the ticket awaiting your installing a
"supported filesystem")
it didn't look like XFS was a terribly good bet. Those issues were
certainly causing concern a couple of years ago.
Faster "raw performance" isn't much good if it comes with a risk of:
- Losing data
- Losing support from vendors
If XFS now *is* getting support from both the development and support
perspectives, then the above concerns may have been invalidated. It
would be very encouraging, if so.
--
output = ("cbbrowne" "@" "gmail.com")
Rules of the Evil Overlord #228. "If the hero claims he wishes to
confess in public or to me personally, I will remind him that a
notarized deposition will serve just as well."
From | Date | Subject | |
---|---|---|---|
Next Message | Samuel Gendler | 2010-10-12 16:23:06 | Re: Slow count(*) again... |
Previous Message | Scott Carey | 2010-10-12 16:02:39 | Re: Slow count(*) again... |
From | Date | Subject | |
---|---|---|---|
Next Message | Chris Browne | 2010-10-12 16:13:09 | Re: large dataset with write vs read clients |
Previous Message | Scott Carey | 2010-10-12 16:02:39 | Re: Slow count(*) again... |