Persistent storage on GPU

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Persistent storage on GPU

elafrit
Hi,

Is there any way to hold persistent storage of data on the gpu so that we don't have to repeatedly the data sent each time.

If I do the next for example :

a_gpu = gpuarray.to_gpu(a)
b_gpu = gpuarray.to_gpu(b)
c_gpu = gpuarray.to_gpu(c)
temp_gpu = culinalg.dot(a_gpu, b_gpu)
d_gpu = culinalg.dot(temp_gpu, c_gpu)

To get the var d_gpu wa the matrix temp_gpu resent to the GPU or the result that already exists on the GPU was used?

thaks.
Reply | Threaded
Open this post in threaded view
|

Re: Persistent storage on GPU

Andreas Kloeckner
Hi,

On Tue, 8 Feb 2011 07:03:46 -0800 (PST), elafrit <[hidden email]> wrote:

> Is there any way to hold persistent storage of data on the gpu so that we
> don't have to repeatedly the data sent each time.
>
> If I do the next for example :
>
> a_gpu = gpuarray.to_gpu(a)
> b_gpu = gpuarray.to_gpu(b)
> c_gpu = gpuarray.to_gpu(c)
> temp_gpu = culinalg.dot(a_gpu, b_gpu)
> d_gpu = culinalg.dot(temp_gpu, c_gpu)
>
> To get the var d_gpu wa the matrix temp_gpu resent to the GPU or the result
> that already exists on the GPU was used?
Of course--transfers are only performed where you say 'to_gpu' or
'.get()'.

Andreas

_______________________________________________
PyCUDA mailing list
[hidden email]
http://lists.tiker.net/listinfo/pycuda

attachment0 (195 bytes) Download Attachment