Skip site navigation (1) Skip section navigation (2)

Re: BBU Cache vs. spindles

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Rob Wultsch <wultsch(at)gmail(dot)com>
Cc: Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, Greg Smith <greg(at)2ndquadrant(dot)com>, jd(at)commandprompt(dot)com, Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com>, Bruce Momjian <bruce(at)momjian(dot)us>, Steve Crawford <scrawford(at)pinpointresearch(dot)com>, pgsql-performance(at)postgresql(dot)org, Ben Chobot <bench(at)silentmedia(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Subject: Re: BBU Cache vs. spindles
Date: 2010-10-26 14:25:38
Message-ID: AANLkTikPs5Um=9LkOS0FaShwrcrpA6qVs5NX=FEfHZFV@mail.gmail.com (view raw or flat)
Thread:
Lists: pgsql-performancepgsql-www
On Tue, Oct 26, 2010 at 10:13 AM, Rob Wultsch <wultsch(at)gmail(dot)com> wrote:
> The double write buffer is one of the few areas where InnoDB does more
> IO (in the form of fsynch's) than PG. InnoDB also has fuzzy
> checkpoints (which help to keep dirty pages in memory longer),
> buffering of writing out changes to secondary indexes, and recently
> tunable page level compression.

Baron Schwartz was talking to me about this at Surge.  I don't really
understand how the fuzzy checkpoint stuff works, and I haven't been
able to find a good description of it anywhere.  How does it keep
dirty pages in memory longer?  Details on the other things you mention
would be interesting to hear, too.

> Given that InnoDB is not shipping its logs across the wire, I don't
> think many users would really care if it used the double writer or
> full page writes approach to the redo log (other than the fact that
> the log files would be bigger). PG on the other hand *is* pushing its
> logs over the wire...

So how is InnoDB doing replication?  Is there a second log just for that?

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Responses

pgsql-performance by date

Next:From: Kevin GrittnerDate: 2010-10-26 14:27:49
Subject: Re: CPUs for new databases
Previous:From: Rob WultschDate: 2010-10-26 14:13:53
Subject: Re: BBU Cache vs. spindles

pgsql-www by date

Next:From: Rob WultschDate: 2010-10-27 04:41:55
Subject: Re: BBU Cache vs. spindles
Previous:From: Rob WultschDate: 2010-10-26 14:13:53
Subject: Re: BBU Cache vs. spindles

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group