Lightweight Task Offloading Exploiting MPI Wait Times for Parallel Adaptive Mesh Refinement

Samfass, Philipp; Weinzierl, Tobias; Charrier, Dominic E.; Bader, Michael

Abstract

Balancing the workload of sophisticated simulations is inherently difficult, since we have to balance both computational workload and memory footprint over meshes that can change any time or yield unpredictable cost per mesh entity, while modern supercomputers and their interconnects start to exhibit fluctuating performance. We propose a novel lightweight balancing technique for MPI+X to accompany traditional, prediction‐based load balancing. It is a reactive diffusion approach that uses online measurements of MPI idle time to migrate tasks temporarily from overloaded to underemployed ranks. Tasks are deployed to ranks which otherwise would wait, processed with high priority, and made available to the overloaded ranks again. This migration is nonpersistent. Our approach hijacks idle time to do meaningful work and is totally nonblocking, asynchronous and distributed without a global data view. Tests with a seismic simulation code developed in the ExaHyPE engine uncover the method's potential. We found speed‐ups of up to 2‐3 for ill‐balanced scenarios without logical modifications of the code base and show that the strategy is capable to react quickly to temporarily changing workload or node performance.

Citation

Samfass, P., Weinzierl, T., Charrier, D. E., & Bader, M. (2020). Lightweight Task Offloading Exploiting MPI Wait Times for Parallel Adaptive Mesh Refinement. Concurrency and Computation: Practice and Experience, 32(24), Article e5916. https://doi.org/10.1002/cpe.5916

• Journal Article Type: Article
• Acceptance Date: May 18, 2020
• Online Publication Date: Jul 9, 2020
• Publication Date: Dec 25, 2020
• Deposit Date: May 18, 2020
• Publicly Available Date: Jul 17, 2020
• Journal: Concurrency and Computation: Practice and Experience
• Print ISSN: 1532-0626
• Electronic ISSN: 1532-0634
• Publisher: Wiley
• Peer Reviewed: Peer Reviewed
• Volume: 32
• Issue: 24
• Article Number: e5916
• DOI: https://doi.org/10.1002/cpe.5916
• Public URL: https://durham-repository.worktribe.com/output/1301915
• Related Public URLs: https://arxiv.org/abs/1909.06096