[IPOL discuss] Warning for the use of a 32-core ?

Nicolas Limare nicolas.limare at cmla.ens-cachan.fr
Fri May 27 12:08:22 CEST 2011

> I just tought about something about the 32 core which is used for
> IPOL. Apparently it's not done yet, but wouldn't it be better to
> warn users that demos are running on a 32-core, and that they have
> to expect a worse performance at home, especially on parallelized
> algorithms?

I think this warning would be good.

If we hardcode it in the demo system we can easily update this
information for all the demos, but for most of the durrent algorithms
there is no difference between 1 core and 32 cores because the code is
not parallel. I think this warning should be left to the authors
initiative; if they expect significantly better performances on the
IPOL server than on a typical home desktop machine, they can inform
the visitor in the "source code" and "online demo" sections of the
algorithm web page.

Nicolas LIMARE - CMLA - ENS Cachan    http://www.cmla.ens-cachan.fr/~limare/
IPOL - image processing on line                          http://www.ipol.im/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://tools.ipol.im/mailman/archive/discuss/attachments/20110527/850f4521/attachment.pgp>

More information about the discuss mailing list