From: Eric Bavier <ericbavier@centurylink.net>
To: Ricardo Wurmus <ricardo.wurmus@mdc-berlin.de>
Cc: guix <guix-devel@gnu.org>
Subject: Re: RFC: building numpy against OpenBLAS.
Date: Fri, 22 May 2015 11:16:00 -0400 (EDT) [thread overview]
Message-ID: <1671511688.83740600.1432307760239.JavaMail.root@centurylink.net> (raw)
In-Reply-To: <idjwq00hd20.fsf@bimsb-sys02.mdc-berlin.net>
----- Ricardo Wurmus <ricardo.wurmus@mdc-berlin.de> wrote:
> python-numpy currently depends on Atlas, which means that it cannot be
> substituted with a binary built elsewhere. OpenBLAS is an alternative
> to Atlas and the binary can be used on all supported CPUs at runtime.
> This makes it possible for us to make numpy substitutable.
[...]
> Anyway, I just wanted to post this here to ask for opinions. Maybe this
> is a bad idea. (In my case it makes sense not to use Atlas, because the
> compile-time tuning is useless when a shared store is used and clients
> use Atlas on machines other than the build host.)
>
I would very much like to see OpenBLAS substituted for atlas whereever possible. The runtime performance of OpenBLAS is quite a bit better than ATLAS. I've seen e.g. OpenBLAS to be roughly 20% faster than ATLAS for GEMM calls on Sandybridge platforms.
`~Eric
next prev parent reply other threads:[~2015-05-22 15:16 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-22 15:00 RFC: building numpy against OpenBLAS Ricardo Wurmus
2015-05-22 15:16 ` Eric Bavier [this message]
2015-06-01 13:11 ` Ricardo Wurmus
2015-06-05 13:52 ` Mark H Weaver
2015-06-05 13:54 ` Mark H Weaver
2015-06-05 15:36 ` Mark H Weaver
2015-06-05 16:01 ` Ricardo Wurmus
2015-06-05 17:33 ` Mark H Weaver
2015-06-07 19:46 ` Ludovic Courtès
2015-06-07 21:32 ` Mark H Weaver
-- strict thread matches above, loose matches on Subject: below --
2015-05-23 7:20 Federico Beffa
2015-05-27 15:50 ` Ricardo Wurmus
2015-05-28 6:38 ` Federico Beffa
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://guix.gnu.org/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1671511688.83740600.1432307760239.JavaMail.root@centurylink.net \
--to=ericbavier@centurylink.net \
--cc=guix-devel@gnu.org \
--cc=ricardo.wurmus@mdc-berlin.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/guix.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).