Difference between revisions of "Streams and Multi-GPU using CUVI"
Line 37: | Line 37: | ||
//Number of data chunks | //Number of data chunks | ||
int streamCount = | int streamCount = 3; | ||
//Creating a 3-channel Image container on GPU | //Creating a 3-channel Image container on GPU | ||
Line 47: | Line 44: | ||
CuviImage* gout = new CuviImage(size,img->depth,1); | CuviImage* gout = new CuviImage(size,img->depth,1); | ||
//Height of each chunk in the image | //Height of each chunk in of the image | ||
size_t cHeight = gimg->height / streamCount; | size_t cHeight = gimg->height / streamCount; | ||
Line 56: | Line 53: | ||
//Chunks sizes for | //Chunks sizes for each stream in bytes | ||
//gpuChunk and hostChunk are mostly different because their pitch is different | //gpuChunk and hostChunk are mostly different because their pitch is different | ||
size_t gpuChunk = cHeight * gimg->pitch, | size_t gpuChunk = cHeight * gimg->pitch, | ||
Line 65: | Line 62: | ||
//Uploading image data to GPU in streams | //Uploading image data to GPU in streams | ||
for(int i=0; i<streamCount; i++) | for(int i=0; i<streamCount; i++) | ||
inputImage->upload(gimg->data + i*gpuChunk, host_img->imageData+ i*hostChunk, host_img->widthStep,streams[i]); | inputImage->upload(gimg->data + i*gpuChunk, host_img->imageData + i*hostChunk, host_img->widthStep, streams[i]); | ||
//Function call on each stream | //Function call on each stream | ||
Line 73: | Line 70: | ||
//Downloading resultant data back to host in streams | //Downloading resultant data back to host in streams | ||
for(int i=0; i<streamCount; i++) | for(int i=0; i<streamCount; i++) | ||
gout->download(host_out->imageData+ i*hostOutChunk , gout + i*gpuOutChunk, host_out->widthStep, streams[i]); | gout->download(host_out->imageData + i*hostOutChunk , gout + i*gpuOutChunk, host_out->widthStep, streams[i]); | ||
//Don't forget to destroying streams and free memory | //Don't forget to destroying streams and free memory |
Revision as of 15:39, 4 May 2012
Using Streams with CUVI
CUVI framework provides a way to use streams with minimal coding effort. Each function call in CUVI takes an optional parameter to specify the stream on which it should run. The code below shows how a simple function call of CUVI can be divided into streaming calls on GPU. For most of the cases this will result in better performance as copying image data to GPU and processing that data on GPU happens simultaneously.
CUVI example
In this example we use CUVI's RGB2Gray function from Color Operations module on a full HD input image
|
Same example with Streams
Streams greatly improve performance of your application by hiding data processing time in data copying time. Instead of waiting for the complete image to be copied on GPU before processing, streaming enables processing the data as it arrives on GPU. Here's how you can use streaming in your application using CUVI:
|