General Matrix Multiply
This paper details the implementation of out of order queues, an OpenCL™ construct that allows independent kernels to execute simultaneously whenever possible, and thus keep all GPU assets fully utilized.
Please see the new portal for OpenCL™ deployments prior to accessing this legacy content.
Learn core concepts of developing OpenCL™ applications with Intel® SDK for OpenCL™ Applications 2019.
Download PDF (1.5 MB)
Download the pdf version of the article
Obtain runtimes to execute or develop OpenCL™ applications on Intel® processors.