A speculative and adaptive MPI rendezvous protocol over RDMA-enabled interconnects

DOI10.1007/S10766-009-0094-9MaRDI QIDQ842850zbMATH OpenOpenAlexFDO

Authors Mohammad J. Rashti, Ahmad Afsahi

Publication date 25 September 2009

Published in International Journal of Parallel Programming (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s10766-009-0094-9

zbMATH Keywords

overlapcommunication progresshigh-performance interconnectsMPI rendezvous protocolRDMA

Mathematics Subject Classification ID

Network design and communication in computer systems (68M10) Network protocols (68M12)

Recommendations

• scientific article; zbMATH DE number 1844529
• High performance RDMA-based MPI implementation over InfiniBand
• scientific article; zbMATH DE number 1844521
• scientific article; zbMATH DE number 1859622
• scientific article; zbMATH DE number 1948474

Cites work

• scientific article; zbMATH DE number 1948423 (Why is no real title available?)
• A pipelined schedule to minimize completion time for loop tiling with computation and communication overlapping

Cited in

(2)

• High performance RDMA-based MPI implementation over InfiniBand
• A fast RDMA offload method for unreliable interconnection networks

Describes a project that uses

Uses Software

• NAS Parallel Benchmarks
• MPICH

This page was built for publication: A speculative and adaptive MPI rendezvous protocol over RDMA-enabled interconnects

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q842850)