Dependency-aware task-based parallel programming models have proven to be successful for developing application software for multicore-based computer architectures. Here we consider the problem of scheduling tasks not only with respect to their inter-dependencies, but also with respect to their usage of resources such as memory and bandwidth. At the software level, this is achieved by user annotations of the task resource consumption. In the run-time system, the annotations are translated into scheduling constraints. Experimental results demonstrating performance gains both for model examples and real applications are presented.

Peer Reviewed