Towards millions of communicating threads
- 25 September 2016
- conference paper
- conference paper
- Published by Association for Computing Machinery (ACM) in Proceedings of the 23rd European MPI Users' Group Meeting
Abstract
We explore in this paper the advantages that accrue from avoiding the use of wildcards in MPI. We show that, with this change, one can efficiently support millions of concurrently communicating light-weight threads using send-receive communication.Keywords
This publication has 33 references indexed in Scilit:
- Designing Scalable Graph500 Benchmark with Hybrid MPI+OpenSHMEM Programming ModelsLecture Notes in Computer Science, 2013
- Unbalanced tree search on a manycore system using the GPI programming modelComputer Science - Research and Development, 2011
- Fine-Grained Multithreading Support for Hybrid Threaded MPI ProgrammingThe International Journal of High Performance Computing Applications, 2010
- UTS: An Unbalanced Tree Search BenchmarkPublished by Springer Science and Business Media LLC ,2007
- High Performance RDMA-Based MPI Implementation over InfiniBandInternational Journal of Parallel Programming, 2004
- Thread Scheduling for Multiprogrammed MultiprocessorsTheory of Computing Systems, 2001
- Compile/run-time support for threaded MPI execution on multiprogrammed shared memory machinesACM SIGPLAN Notices, 1999
- Linearizability: a correctness condition for concurrent objectsACM Transactions on Programming Languages and Systems, 1990
- Communicating Sequential ProcessesPublished by Springer Science and Business Media LLC ,1978
- Space/time trade-offs in hash coding with allowable errorsCommunications of the ACM, 1970