Postgresql on file system EXT2 or EXT3

[Andrew Sullivan]

Re: Postgresql on file system EXT2 or EXT3

On Fri, Nov 28, 2003 at 02:46:50PM -0800, Mike Benoit wrote:[color=blue]
> I would be very interested in seeing actual PGBENCH results with
> databases on the different file systems, thats the only way you will
> know for sure which file system is best for the task.[/color]

I suspect because of the nature of its workload, in my experience
pg_bench is lousy for measuring filesystem performance: it always
bottlenecks somewhere else.

Chris Browne did some work for us some time ago evaluating XFS, JFS,
and ext3, and concluded that JFS was the best under a high-update
load; that workload was selected precisely because it was I/O bound.
I thought he sent the results to the -performance list, but I can't
put my hands on the email right now. Chris?

A

--
----
Andrew Sullivan 204-4141 Yonge Street
Afilias Canada Toronto, Ontario Canada
<andrew@liberty rms.info> M2P 2A8
+1 416 646 3304 x110


---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster

[Christopher Browne]

Re: Postgresql on file system EXT2 or EXT3

Centuries ago, Nostradamus foresaw when andrew@libertyr ms.info (Andrew Sullivan) would write:[color=blue]
> On Fri, Nov 28, 2003 at 02:46:50PM -0800, Mike Benoit wrote:[color=green]
>> I would be very interested in seeing actual PGBENCH results with
>> databases on the different file systems, thats the only way you
>> will know for sure which file system is best for the task.[/color]
>
> I suspect because of the nature of its workload, in my experience
> pg_bench is lousy for measuring filesystem performance: it always
> bottlenecks somewhere else.
>
> Chris Browne did some work for us some time ago evaluating XFS, JFS,
> and ext3, and concluded that JFS was the best under a high-update
> load; that workload was selected precisely because it was I/O
> bound. I thought he sent the results to the -performance list, but I
> can't put my hands on the email right now. Chris?[/color]

Yes, that's right, and the paucity of hard-and-fast details comes from
the fact that the sample workload was, well, pretty proprietary. It
was an honest-to-goodness real workload for one of the registries,
which means that I can't give out copies. (With suitable caveats of
"or else I'd have to kill you," or, more realistically "or else they'd
have to kill me..." :-(.)

The results repeated well, with JFS being ~20% faster than ext3 or
XFS. (I found XFS marginally slower for this benchmark than ext3, but
the difference was small enough that I wouldn't trust that as a True
Conclusion.)

The actual measurements are probably in the internal Systems archives;
I am generally disinclined to give out numbers publicly, in view of
the public unavailability of the workload.

What I had previously reported was actually on the pgsql-admin list...
<http://archives.postgr esql.org/pgsql-admin/2003-09/msg00284.php>
--
let name="aa454" and tld="freenet.ca rleton.ca" in String.concat "@" [name;tld];;

http://cbbrowne.com/info/sap.html

REALITY is a policy phased out early in the Eisenhower administration.

[Mike Benoit]

Re: Postgresql on file system EXT2 or EXT3

On Sat, 2003-11-29 at 02:07, Unihost Web Hosting wrote:[color=blue]
> The next fs I install will be XFS after much deliberation and
> conversation with like minded fellow compugeeks, since it's at least as
> good as any other journaling fs, but has the added bonus that
> filesystems can be *grown* without the aid of LVM, etc. Which would be
> a huge bonus.
>[/color]

resize_reiserfs
resize2fs

We grow a EXT3 file system all the time on a Linear RAID array. Works
fine without LVM. Shrinking can be a different issue though.
[color=blue]
> Just my 2 cents.
>
> Tony.
>
> Joshua D. Drake wrote:
>[color=green]
> > | Don't go on EXT2, its not reliable and takes lots of time to start
> > after an
> >
> > Actually EXT2 is quite reliable and it is also quite fast. However your
> > point is accurate about start up time after a crash.
> >[color=darkred]
> >> The most promising FS is Reiserfs v4
> >> http://www.namesys.com/v4/v4.html
> >>
> >>
> >>[/color]
> > Although Reiser is promising, I wouldn't touch it. It is beta, frankly my
> > experience is that even their stable stuff is still beta.
> >
> > If you want a native, reliable, stable FS for Linux. Use JFS or XFS
> > (when 2.6 comes out)
> >
> > Sincerely,
> >
> > Joshua D. Drake
> >
> >
> >
> >
> >
> >
> >
> >[color=darkred]
> >> If you cant wait I suggest XFS or JFS.
> >>
> >> Look in the archives for all the explanations.
> >>
> >> Ohhh, and don't use IDE Drives, only SCSI.
> >>
> >> Cheer
> >> --------------------------
> >> Canaan Surfing Ltd.
> >> Internet Service Providers
> >> Ben-Nes Michael - Manager
> >> Tel: 972-4-6991122
> >> Fax: 972-4-6990098
> >> http://www.canaan.net.il
> >> --------------------------
> >> ----- Original Message ----- From: "Carmen Wai" <wai_carmen@hot mail.com>
> >> To: <pgsql-general@postgre sql.org>
> >> Sent: Thursday, November 27, 2003 5:00 PM
> >> Subject: [GENERAL] Postgresql on file system EXT2 or EXT3
> >>
> >>
> >>
> >>
> >>> Hello:
> >>>
> >>> I would like to know whether there is any different in installing
> >>>
> >>
> >> Postgresql
> >>
> >>
> >>> on the Linux system with file system of EXT2 or EXT3. I have two
> >>> machines
> >>> with idential OS (Red Hat 7.3 install with postgresql 7.3.4) but with
> >>> different file system, 1 is EXT2 and the other is EXT3. When I insert
> >>>
> >>
> >> 10,000
> >>
> >>
> >>> records to the two machines, I found that the machine with EXT2 insert
> >>>
> >>
> >> much
> >>
> >>
> >>> quicker than the other with EXT3.
> >>>
> >>> Is postgresqk perform better with EXT2 file system?
> >>>
> >>> Thanks a lot!
> >>> Carmen
> >>>
> >>> _______________ _______________ _______________ _______________ _____
> >>> Linguaphone : Learning English? Get Japanese lessons for FREE
> >>> http://go.msnserver.com/HK/30476.asp
> >>>
> >>>
> >>> ---------------------------(end of
> >>> broadcast)---------------------------
> >>> TIP 7: don't forget to increase your free space map settings
> >>>
> >>>
> >>
> >>
> >>
> >>
> >> ---------------------------(end of broadcast)---------------------------
> >> TIP 1: subscribe and unsubscribe commands go to majordomo@postg resql.org
> >>
> >>[/color]
> >[/color]
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
> (send "unregister YourEmailAddres sHere" to majordomo@postg resql.org)[/color]
--
Best Regards,

Mike Benoit
NetNation Communications Inc.
Systems Engineer
Tel: 604-684-6892 or 888-983-6600
---------------------------------------

Disclaimer: Opinions expressed here are my own and not
necessarily those of my employer


---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo@postg resql.org