Communication lower bounds for distributed-memory matrix multiplication

DOI10.1016/J.JPDC.2004.03.021MaRDI QIDQ1886368zbMATH OpenOpenAlexDBLPWikidataFDO

Authors Dror Irony, Sivan Toledo, Alexander Tiskin

Publication date 18 November 2004

Published in Journal of Parallel and Distributed Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.jpdc.2004.03.021

zbMATH Keywords

Lower boundsMatrix multiplicationCommunicationDistributed memory

Mathematics Subject Classification ID

Complexity and performance of numerical algorithms (65Y20) Distributed algorithms (68W15)

Recommendations

• Graph expansion and communication costs of fast matrix multiplication
• Minimizing communication in numerical linear algebra
• scientific article
• Parallel matrix multiplication: a systematic journey
• Parallel complexity of matrix multiplication

Cited in

(25)

• REVISITING MATRIX PRODUCT ON MASTER-WORKER PLATFORMS
• Massively parallel sparse matrix function calculations with NTPoly
• HPMaX: heterogeneous parallel matrix multiplication using CPUs and GPUs
• Exploiting multiple levels of parallelism in sparse matrix-matrix multiplication
• Parallel complexity of matrix multiplication
• Parallel matrix multiplication: a systematic journey
• Communication Lower Bounds and Optimal Algorithms for Multiple Tensor-Times-Matrix Computation
• A cache-optimal alternative to the unidirectional hierarchization algorithm
• Graph expansion and communication costs of fast matrix multiplication
• Communication efficient matrix multiplication on hypercubes
• Numerical algorithms for high-performance computational science
• Cache optimization and performance modeling of batched, small, and rectangular matrix multiplication on Intel, AMD, and Fujitsu processors
• Task-based parallel programming for scalable matrix product algorithms
• Algorithm 953: Parallel library software for the multishift QR algorithm with aggressive early deflation
• Introduction to communication avoiding algorithms for direct methods of factorization in linear algebra
• A bridging model for multi-core computing
• Matrix exponentials and parallel prefix computation in a quantum control problem
• Pebbling Game and Alternative Basis for High Performance Matrix Multiplication
• scientific article; zbMATH DE number 6691438 (Why is no real title available?)
• Communication lower bounds and optimal algorithms for numerical linear algebra
• Communication lower bounds of bilinear algorithms for symmetric tensor contractions
• Distributed control for large-scale systems with adaptive event-triggering
• On the cost of iterative computations
• Oblivious algorithms for multicores and networks of processors
• Parallel time integration using batched BLAS (Basic Linear Algebra Subprograms) routines

Describes a project that uses

Uses Software

• PHiPAC
• SUMMA
• GEMM
• ATLAS

This page was built for publication: Communication lower bounds for distributed-memory matrix multiplication

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1886368)