Chunking a gpuArray - can it be done?
信息
此问题已关闭。 请重新打开它进行编辑或回答。
显示 更早的评论
My question is a little bit theoretical. I'm trying to develop a code to process a signal, therefore using concepts of DSP. Working with gpuArray is great, but I still need to optimize my code to run even faster. I know that MATLAB does the multi-threading intrinsically, but my question is: if I split my signal (it has 0.5-1GB) into smaller arrays (let's say, 1024 samples each, making it a block, so to speak), would it be faster? Or would it be the same if I tried to parallelized this way?
Thanks in advance!
1 个评论
Edric Ellis
2015-11-9
Could you post a simple example of the change you're considering making?
回答（0 个）
此问题已关闭。
关闭：
2021-8-20
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
