A bandwidth latency tradeoff for broadcast and reduction

DOI10.1016/S0020-0190(02)00473-8zbMath1173.68847DBLPjournals/ipl/SandersS03OpenAlexW1982169443WikidataQ60307050 ScholiaQ60307050MaRDI QIDQ1007568

Jop F. Sibeyn, Peter Sanders

Publication date: 23 March 2009

Published in: Information Processing Letters (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/s0020-0190(02)00473-8

zbMATH Keywords

treereductionparallel algorithmsmeshbroadcastcollective communicationfull-duplexhalf-duplexhierarchical crossbarsingle ported

Mathematics Subject Classification ID

Parallel algorithms in computer science (68W10)

Related Items (2)

Techniques for pipelined broadcast on Ethernet switched clusters ⋮ Optimal broadcast for fully connected processor-node networks

Uses Software

• MPI

Cites Work

• Unnamed Item
• Optimal and near-optimal algorithms for \(k\)-item broadcast
• Broadcasting multiple messages in simultaneous send/receive systems
• Optimum broadcasting and personalized communication in hypercubes

This page was built for publication: A bandwidth latency tradeoff for broadcast and reduction