Encapsulating Multiple Communication-Cost Metrics in Partitioning Sparse Rectangular Matrices for Parallel Matrix-Vector Multiplies

DOI10.1137/S1064827502410463zbMath1061.65036OpenAlexW4302764849MaRDI QIDQ4652296

Bora Uçar, Cevdet Aykanat

Publication date: 25 February 2005

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1137/s1064827502410463

zbMATH Keywords

performancenumerical examplesiterative methodparallel computingrectangular matrixmatrix partitioninghypergraph partitioningmessage latencystructurally unsymmetric matrixcommunication hypergraphmatrix-vector multiplymessage volume

Mathematics Subject Classification ID

Hypergraphs (05C65) Graphs and linear algebra (matrices, eigenvalues, etc.) (05C50) Parallel numerical computation (65Y05) Complexity and performance of numerical algorithms (65Y20)

Related Items

Simultaneous Computational and Data Load Balancing in Distributed-Memory Setting, Parallel multilevel algorithms for hypergraph partitioning, Multi-level direct \(K\)-way hypergraph partitioning with multiple constraints and fixed vertices, The Effect of Various Sparsity Structures on Parallelism and Algorithms to Reveal Those Structures, Adapting Iterative-Improvement Heuristics for Scheduling File-Sharing Tasks on Heterogeneous Platforms, A Matrix Partitioning Interface to PaToH in MATLAB, Complex Network Partitioning Using Label Propagation, Network Flow-Based Refinement for Multilevel Hypergraph Partitioning

Uses Software

• METIS
• LAM-MPI
• Chaco
• PaToH