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

Performance tests with parallel gmake

From: Bruce Momjian <bruce(at)momjian(dot)us>
To: PostgreSQL-development <pgsql-hackers(at)postgreSQL(dot)org>
Subject: Performance tests with parallel gmake
Date: 2010-11-27 21:18:01
Message-ID: 201011272118.oARLI1q02199@momjian.us (view raw or flat)
Thread:
Lists: pgsql-hackers
I ran some tests on my dual-Xeon machine to see what speedup using
parallel gmake.  After two builds to warm up the cache, I saw when
alternating gmake and gmake -j2:

gmake           0m57.207s
gmake -j2       0m44.305s
gmake           0m58.691s
gmake -j2       0m33.979s

As you can see, gmake -j2 is ~35% faster than gmake without -j. I will
be using -j from now on.  -j3 shaves another second off.

FYI, I also use ccache for compiles.

-- 
  Bruce Momjian  <bruce(at)momjian(dot)us>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

  + It's impossible for everything to be true. +

pgsql-hackers by date

Next:From: Bruce MomjianDate: 2010-11-27 22:30:11
Subject: Re: What do these terms mean in the SOURCE CODE?
Previous:From: Tom LaneDate: 2010-11-27 21:11:17
Subject: Re: Small documentation enhancement for default_tablespace

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