[asterisk-dev] nVidia Cuda

[Wai Wu]

I have to agree with Matthew here. The widely available GPUs is one of the reasons I got into this. If you look at the architecture of the G80, it is not bad at all. I also agree that becasue of the nature of graphic processing, GPUs in general tend not to have as many processor as they clame to have, but that will change. Also, when porting voice applications to these GPU, we are not look to Parallelize the code, instead, we are look to the same thing for many channel. I think GPUs are perfect for servers such as asterisk where FFT, mixing, encoding, and decoding are performed on all channels all the time.

-----Original Message-----
From: asterisk-dev-bounces at lists.digium.com on behalf of Matthew Rubenstein
Sent: Mon 3/5/2007 2:38 PM
To: Tzafrir Cohen
Cc: Asterisk-Dev
Subject: Re: [asterisk-dev] nVidia Cuda
 


> My personal feeling is that the price of the GPU would be better spent
> on a stonger CPU. Not to mention the fact that the GPU architecture
> seem
> to scale very badly on a standard CPU system. But I'd be gladly proven
> wrong.

	Not only are GPUs cheap compared with CPUs per MFLOPS (GFLOPS), but
multiple GPUs can be run in a single host across a PCI-e bus. The
compute parallelism against parallel streams inside the low-latency,
high-bandwidth bus is scalable in ways not possible with the CPU. Adding
GPUs doesn't require adding the rest of the host HW, like power supply,
IO, chassis, etc that can be shared by the rest of the machine. What
makes you say that "the GPU architecture seem[s] to scale very badly on
[a] standard CPU system"?


-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 3415 bytes
Desc: not available
Url : http://lists.digium.com/pipermail/asterisk-dev/attachments/20070305/357ade90/winmail.bin