Callback-based completion notification using MPI Continuations

作者: George Bosilca , Christoph Niethammer , José Gracia , Joseph Schuchart , Philipp Samfass

DOI: 10.1016/J.PARCO.2021.102793

关键词:

摘要: Abstract Asynchronous programming models (APM) are gaining more and traction, allowing applications to expose the available concurrency a runtime system tasked with coordinating execution. While MPI has long provided support for multi-threaded communication non-blocking operations, it falls short of adequately supporting APMs as correctly efficiently handling in different is still challenge. We have previously proposed an extension standard providing operation completion notifications using callbacks, so-called Continuations. This interface flexible enough accommodate wide range APMs. In this paper, we present described that allows finer control behavior Continuations interface. then some our first experiences context applications, including NAS parallel benchmarks, PaRSEC task-based system, load-balancing scheme within adaptive mesh refinement solver called ExaHyPE. show interface, implemented inside Open MPI, enables low-latency, high-throughput outperform solutions application space.

参考文章(43)
Howard Pritchard, Igor Gorodetsky, Darius Buntinas, A uGNI-based MPICH2 nemesis network module for the cray XE EuroMPI'11 Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface. pp. 110- 119 ,(2011) , 10.1007/978-3-642-24449-0_14
Daniel P. Friedman, Christopher T. Haynes, Eugene Kohlbecker, Programming with Continuations Program Transformation and Programming Environments. pp. 263- 274 ,(1984) , 10.1007/978-3-642-46490-4_23
Robert Latham, William Gropp, Robert Ross, Rajeev Thakur, Extending the MPI-2 generalized request interface PVM/MPI'07 Proceedings of the 14th European conference on Recent Advances in Parallel Virtual Machine and Message Passing Interface. ,vol. 4757, pp. 223- 232 ,(2007) , 10.1007/978-3-540-75416-9_33
Torsten Hoefler, Greg Bronevetsky, Brian Barrett, Bronis R. de Supinski, Andrew Lumsdaine, Efficient MPI Support for Advanced Hybrid Programming Models Recent Advances in the Message Passing Interface. pp. 50- 61 ,(2010) , 10.1007/978-3-642-15646-5_6
J. Choi, J.J. Dongarra, R. Pozo, D.W. Walker, ScaLAPACK: a scalable linear algebra library for distributed memory concurrent computers symposium on frontiers of massively parallel computation. pp. 120- 127 ,(1992) , 10.1109/FMPC.1992.234898
James Dinan, Pavan Balaji, David Goodell, Douglas Miller, Marc Snir, Rajeev Thakur, Enabling MPI interoperability through flexible communication endpoints Proceedings of the 20th European MPI Users' Group Meeting on - EuroMPI '13. pp. 13- 18 ,(2013) , 10.1145/2488551.2488553
Hartmut Kaiser, Thomas Heller, Bryce Adelstein-Lelbach, Adrian Serio, Dietmar Fey, HPX: A Task Based Programming Model in a Global Address Space Proceedings of the 8th International Conference on Partitioned Global Address Space Programming Models. pp. 6- ,(2014) , 10.1145/2676870.2676883
George Bosilca, Aurelien Bouteiller, Anthony Danalis, Thomas Herault, Pierre Lemarinier, Jack Dongarra, DAGuE: A Generic Distributed DAG Engine for High Performance Computing ieee international symposium on parallel & distributed processing, workshops and phd forum. pp. 1151- 1158 ,(2011) , 10.1109/IPDPS.2011.281