Featured Article


Port Some CUDA Codes To Intel Xeon Phi Simply and Efficiently

This tutorial shows that it relatively easy to port many CUDA C/C++ source codes to OpenMP. In the past, such efforts were not generally considered worthwhile because of the large performance difference between multicore processors (that use OpenMP) and GPUs. The introduction of teraflop/s Intel Xeon Phi coprocessors eliminated that performance difference, which makes it much more […]

Featured News, Commentary, and Analysis

More Posts from this Category

Featured Tutorials

More Posts from this Category