[asterisk-dev] nVidia Cuda

[Tzafrir Cohen]

On Mon, Mar 05, 2007 at 05:10:47PM -0500, Wai Wu wrote:
> 
> 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.

If this is so generic, please port the standard speex library code to
use the GPU as a backend. 

If not: pick a simple single-threaded program that encodes speex (I
suggested sox), port it to use the GPU and show first how to run it
there. Then formore that one of those together. By then you should
probably have a much better idea of the expected performance.

If this works well, start working on integrating it with Asterisk (or
OpenSer, or sipX, or whatever).

-- 
               Tzafrir Cohen       
icq#16849755                    jabber:tzafrir at jabber.org
+972-50-7952406           mailto:tzafrir.cohen at xorcom.com       
http://www.xorcom.com  iax:guest at local.xorcom.com/tzafrir