Re: [HACKERS] Performance while loading data and indexing - Mailing list pgsql-general

From Mats Lofkvist
Subject Re: [HACKERS] Performance while loading data and indexing
Date
Msg-id y2q1y7f3ecy.fsf@algonet.se
Whole thread Raw
In response to Re: [HACKERS] Performance while loading data and indexing  (Neil Conway <neilc@samurai.com>)
List pgsql-general
neilc@samurai.com (Neil Conway) writes:

[snip]
> > Well, I have experienced data loss from ext2 before.  Also, recovery
> > from crashes on large file systems take a very, very long time.
>
> Yes, but wouldn't you face exactly the same issues if you ran a
> UFS-like filesystem in asynchronous mode? Albeit it's not the default,
> but performance in synchronous mode is usually pretty poor.
>
> The fact that ext2 defaults to asynchronous mode and UFS (at least on
> the BSDs) defaults to synchronous mode seems like a total non-issue to
> me. Is there any more to the alleged difference in reliability?

UFS on most unix systems (BSD, solaris etc) defaults to sync
metadata, async data which is a mode that is completely missing
from ext2 as far as I know.

This is why UFS is considered safer than ext2. (Running with
'sync' is too slow to be a usable alternative in most cases.)

      _
Mats Lofkvist
mal@algonet.se


PS The BSD soft updates yields the safety of the default sync
   metadata / async data mode while being at least as fast as
   running fully async.

pgsql-general by date:

Previous
From: Andriy Tkachuk
Date:
Subject: Re: query speed depends on lifetime of frozen db?
Next
From: Mats Lofkvist
Date:
Subject: Re: Performance while loading data and indexing