Re: DB logging (was: Problem with the numbers I reported yesterday)

From: ocie(at)paracel(dot)com
To: kgor(at)inetspace(dot)com (Kent S(dot) Gordon)
Cc: jwieck(at)debis(dot)com, ocie(at)paracel(dot)com, maillist(at)candle(dot)pha(dot)pa(dot)us, boersenspiel(at)vocalweb(dot)de, pgsql-hackers(at)postgreSQL(dot)org
Subject: Re: DB logging (was: Problem with the numbers I reported yesterday)
Date: 1998-02-17 01:44:46
Message-ID: 9802170144.AA00326@dolomite.paracel.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Kent S. Gordon wrote:

[SNIP]

> Log files do get very big with image logging. I would not expect a
> huge win in performance unless the log device is a raw device. On a
> cooked device (file system) buffer cache effects are very large (all
> disk data is being buffered both by postgresql and the OS buffer
> cache. The buffer cache is actual harmfully in this case, since data
> is not reused, and the writes are synced. The number of writes to the
> log also flush out other buffer from the cache leading to even more
> io.). If a system does not have raw devices (linux, NT), it would be
^^^^^
What exactly do you mean by "raw devices" that it is not offered by
Linux? If I take a partition of one of my hard drives and I don't
make a filesystem on it, I can perform reads and writes on the "raw
device" /dev/hd?? or /dev/sd?? I didn't think these writes were
buffered (if that's what you're referring to), but I could be wrong.

Ocie Mitchell.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 1998-02-17 03:15:58 Re: DB logging (was: Problem with the numbers I reported yesterday)
Previous Message The Hermit Hacker 1998-02-17 01:31:19 Re: [HACKERS] configure doesn't handle --with-compiler correctly