libo at buaa.edu.cn
Sun Aug 24 08:22:07 EDT 2008
It is about 10 years.
GPU had been focused on the graphics for many years and as we knew that some game consoles going with great computational power comparing to common CPUs, which gave some people the hints to utilize them for general computation. And then to professional graphic cards, and now on general GPUs.
There are two main stream GPGPU architectures at the moment: CUDA and Brook+/CAL, from NVidia and AMD respectively.
CUDA is a very good tools for TLP based GPGPU programming and CAL is a bit close to the hardware.
Hopefully, we can get the Larrabee from Intel soon, which is a GPU or many-core processor with amazing power.
----- Original Message -----
From: "andrew holway" <andrew at moonet.co.uk>
To: "Beowulf" <beowulf at beowulf.org>
Sent: Saturday, August 23, 2008 8:06 AM
Subject: [Beowulf] gpgpu
> How long have people been doing gpgpu for?
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
More information about the Beowulf