Enhancing MPI+OpenMP Task Based Applications for Heterogeneous Architectures with GPU Support

Abstract

Heterogeneous supercomputers are widespread over HPC systems and programming efficient applications on these architectures is a challenge. Task-based programming models are a promising way to tackle this challenge. Since OpenMP 4.0 and 4.5, the target directives enable to offload pieces of code to GPUs and to express it as tasks with dependencies. Therefore, heterogeneous machines can be programmed using MPI+OpenMP(task+target) to exhibit a very high level of concurrent asynchronous operations for which data transfers, kernel executions, communications and CPU computations can be overlapped. Hence, it is possible to suspend tasks performing these asynchronous operations on the CPUs and to overlap their completion with another task execution. Suspended tasks can resume once the associated asynchronous event is completed in an opportunistic way at every scheduling point. We have integrated this feature into the MPC framework and validated it on a AXPY microbenchmark and evaluated on a MPI+OpenMP(tasks) implementation of the LULESH proxy applications. The results show that we are able to improve asynchronism and the overall HPC performance, allowing applications to benefit from asynchronous execution on heterogeneous machines.

Investigating Dependency Graph Discovery Impact on Task-based MPI+OpenMP Applications Performances

“…Recent works [2][3][4] have provided solutions for interoperability issues when nesting MPI communications within OpenMP tasks [5][6][7], enabling the overlap of communications through OpenMP task scheduling. Yet, few applications have migrated towards such a task-based composition: at least two attempts led to convincing performances with the porting of the Cholesky factorization [6] and the hydrodynamics proxy-app LULESH [8]. Other attempts failed to implement functional applications due to interoperability issues [5], had to tinker application codes [7] to sequentialize communications, or added coarse barriers losing potential communication overlap [9].…”