Towards Task-Parallel Reductions in OpenMP
Towards Task-Parallel Reductions in OpenMP. In Proceedings of IWOMP 2015. 15th International Workshop on OpenMP (pp. 189-201). Aachen, GERMANY.
We present the specification of task-parallel reductions and explore issues for programmers and software vendors regarding programming transparency as well as the impact on the current standard with respect to nesting, untied task support and task data dependencies.
Abstract
Reductions represent a common algorithmic pattern in many scientific applications. OpenMP∗ has always supported them on parallel and worksharing constructs. OpenMP 3.0’s tasking constructs enable new parallelization opportunities through the annotation of irregular algorithms. Unfortunately the tasking model does not easily allow the expression of concurrent reductions, which limits the general applicability of the programming model to such algorithms. In this work, we present an extension to OpenMP that supports task-parallel reductions on task and taskgroup constructs to improve productivity and programmability. We present specification of the feature and explore issues for programmers and software vendors regarding programming transparency as well as the impact on the current standard with respect to nesting, untied task support and task data dependencies. Our performance evaluation demonstrates comparable results to hand coded task reductions.