From: tmoran@acm.org
Subject: Re: Multithreaded scientific programming
Date: Wed, 04 Oct 2006 23:18:06 -0500
Date: 2006-10-04T23:18:06-05:00 [thread overview]
Message-ID: <jqOdnSEBdO_jGrnYnZ2dnUVZ_tqdnZ2d@comcast.com> (raw)
In-Reply-To: 1160007272.875375.190830@b28g2000cwb.googlegroups.com
After getting a dual-core (Windows) machine I modified my Ada library
2-D FFT to take a CPU_Count parameter and do the job using that
many separate tasks. Two Ada tasks makes a very pleasing speedup.
Whether 80 would help, and what size cache you might want, I'll
be curious to find out.
next prev parent reply other threads:[~2006-10-05 4:18 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1159978124.458442.80520@m7g2000cwm.googlegroups.com>
[not found] ` <4523e925$1@nntp.unige.ch>
2006-10-04 21:47 ` Multithreaded scientific programming mrohan
2006-10-05 0:14 ` jimmaureenrogers
2006-10-05 4:18 ` tmoran [this message]
2006-10-05 16:32 ` David B. Chorlian
2006-10-05 17:58 ` jimmaureenrogers
2006-10-05 18:35 ` Martin Krischik
2006-10-06 18:22 ` Björn Persson
2006-10-06 20:57 ` jimmaureenrogers
2006-10-09 6:36 ` Martin Krischik
2006-10-05 18:58 ` Francis Burton
2006-10-05 20:20 ` Pascal Obry
2006-10-06 1:28 ` jimmaureenrogers
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox