From: Tom Lane <tgl@sss.pgh.pa.us>
To: "t.dalpozzo@gmail.com" <t.dalpozzo@gmail.com>
cc: pgsql-performance@postgresql.org
Subject: Re: can't explain commit performance win7 vs linux : 8000/s vs 419/s
In-reply-to: <576EAF26.8080508@gmail.com>
References: <576EAF26.8080508@gmail.com>
Comments: In-reply-to "t.dalpozzo@gmail.com" <t.dalpozzo@gmail.com>
	message dated "Sat, 25 Jun 2016 18:19:50 +0200"
Date: Sat, 25 Jun 2016 14:08:35 -0400
Message-ID: <17637.1466878115@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

"t.dalpozzo@gmail.com" <t.dalpozzo@gmail.com> writes:
> Performance:
> Win7: 8000 write/sec
> Linux: 419 write/sec

My immediate reaction to that is that Windows isn't actually writing
the data to disk when it should in order to guarantee that commits
are persistent.  There are multiple layers that might be trying to
optimize away the writes, and I don't know enough about Windows to
help you debug it.  But see

https://www.postgresql.org/docs/9.5/static/wal-reliability.html

for some discussion.

> I don't figure out why such a difference. Also what should I expect? 
> Which one is reasonable?

The lower number sounds a lot more plausible for laptop-grade hardware.
If you weren't using an SSD I wouldn't believe that one was doing
persistent commits either.

			regards, tom lane


-- 
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance