Computing BLAS level-2 operations on workstation clusters using the divisible load paradigm

DOI10.1016/J.MCM.2004.01.004MaRDI QIDQ2486787zbMATH OpenOpenAlexFDO

Authors Debasish Ghose, Hyoung Joong Kim

Publication date 17 August 2005

Published in Mathematical and Computer Modelling (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.mcm.2004.01.004

zbMATH Keywords

BLAS Level-2 operationsDivisible loadsLoad partitioningMatrix-vector productsOptimal load distribution

Mathematics Subject Classification ID

Performance evaluation, queueing, and scheduling in the context of computer systems (68M20)

Recommendations

• Load partitioning and trade-off study for large matrix-vector computations in multicast bus networks with communictaion delays
• Parallel matrix multiplication on networked microcomputers
• scientific article; zbMATH DE number 1203406
• Large matrix-vector products on distributed bus networks with communication delays using the divisible load paradigm: Performance analysis and simulation
• Performance bottlenecks and potentials of parallel computing on networks of workstations

Cites work

• An efficient load distribution strategy for a distributed linear network of processors with communication delays
• Closed form solutions for bus and tree networks of processors load sharing a divisible job
• scientific article; zbMATH DE number 417773 (Why is no real title available?)
• scientific article; zbMATH DE number 4213315 (Why is no real title available?)
• scientific article; zbMATH DE number 1031899 (Why is no real title available?)
• Large matrix-vector products on distributed bus networks with communication delays using the divisible load paradigm: Performance analysis and simulation
• Load partitioning and trade-off study for large matrix-vector computations in multicast bus networks with communictaion delays
• Molecular dynamics and lattice gas parallel algorithms for transputers and networked workstations
• Parallel matrix-vector product on rings with a minimum of communications
• Scheduling divisible loads in a three-dimensional mesh of processors.
• Scheduling divisible loads in bus networks with arbitrary processor release times

Cited in

(3)

• Load-skewing task assignment to minimize communication conflicts on network of workstations
• Performance bottlenecks and potentials of parallel computing on networks of workstations
• Large matrix-vector products on distributed bus networks with communication delays using the divisible load paradigm: Performance analysis and simulation

Describes a project that uses

Uses Software

• BLAS

This page was built for publication: Computing BLAS level-2 operations on workstation clusters using the divisible load paradigm

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2486787)