OpenMP / OpenACC programming


OpenMP has become the de-facto standard for multitasks parallel programming on shared memory systems.

OpenACC is an emerging standard for parallel computing designed to ease the use of accelerators like GPU, or coming many-core processors; OpenACC aims to help programmers to create high-level host+accelerator program by providing directives and libraries to initialize the accelerator, manage data or program transfers between the host and accelerator, or initiate accelerator startup and shutdown. OpenACC is real-wolrd implementation of parallel programming concepts for accelerators that could be part of a future OpenMP standard.


Days 1 and 2 : Introduction to OpenMP, hands-on session

Day 3 : Presentation of the OpenACC programming model;  hands-on session using CAPS/HMPP compiling toolchain

The agenda of this meeting is empty