From: Bruce Momjian <pgman@candle.pha.pa.us>
To: Ron Johnson <ron.l.johnson@cox.net>
Cc: PgSQL Performance ML <pgsql-performance@postgresql.org>
Subject: Re: Perfomance Tuning
Date: Wed, 13 Aug 2003 11:37:16 -0400 (EDT)
Message-ID: <200308131537.h7DFbGu23405@candle.pha.pa.us> (raw)
In-Reply-To: <1060745432.30778.39.camel@haggis>
Ron Johnson wrote:
> On Tue, 2003-08-12 at 13:39, Bruce Momjian wrote:
> > OK, I got some hard evidence. Here is a discussion on the Linux kernel
> > mailing list with postings from Allen Cox (ac Linux kernels) and Stephen
> > Tweedie (ext3 author).
> >
> > http://www.tux.org/hypermail/linux-kernel/1999week14/subject.html#start
> >
> > Search for "softupdates and ext2".
> >
> > Here is the original email in the thread:
> >
> > http://www.tux.org/hypermail/linux-kernel/1999week14/0498.html
> >
> > Summary is at:
> >
> > http://www.tux.org/hypermail/linux-kernel/1999week14/0571.html
> >
> > and conclusion in:
> >
> > http://www.tux.org/hypermail/linux-kernel/1999week14/0504.html
> >
> > I now remember the issue --- ext2 makes all disk changes asynchonously
> > (unless you mount it via sync, which is slow). This means that the file
> > system isn't always consistent on disk.
> >
> > UFS has always sync metadata (file/directory creation) to the disk so
> > the disk was always consistent, but doesn't sync the data to the disk,
> > for performance reasons. With soft updates, the metadata writes are
> > delayed, and written to disk in an order that keeps the file system
> > consistent.
> >
> > Is this enough evidence, or should I keep researching?
>
> This is all 4 years old, though. Isn't that why the ext3 "layer" was
Yes, it is four years old, but no one has told me ext2 has changed in
this regard, and seeing that they created ext3 to fix these aspects, I
would think ext2 hasn't changed.
> created, and filesystems like reiserFS, XFS and (kinda) JFS were added
> to Linux?
Yes, it is those ext2 limitations that caused the development of ext3
and the others. However, they went much father than implementing a
crash-safe file system, but rather enabled a file system that doesn't
need fsck on crash reboot. This causes fsync of data and metadata (file
creation), which slows down the file system, and PostgreSQL doesn't need
it.
You can mount ext3 and others with data=writeback to fsync only
metadata, but it isn't the default.
I am not sure what the ext3 layer is.
--
Bruce Momjian | http://candle.pha.pa.us
pgman@candle.pha.pa.us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients from your mail client; the link fills in
the headers below: reply via email
Some clients, webmail especially, drop In-Reply-To and References from
a link, and the reply then starts a new thread: use the mbox below.
To: pgsql-performance@postgresql.org
Cc: pgman@candle.pha.pa.us, ron.l.johnson@cox.net
Subject: Re: Perfomance Tuning
In-Reply-To: <200308131537.h7DFbGu23405@candle.pha.pa.us>
References: <1060745432.30778.39.camel@haggis> <200308131537.h7DFbGu23405@candle.pha.pa.us>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox