The impact of memory organization on the performance of matrix calculations

DOI10.1016/S0167-8191(05)80116-4zbMATH Open0728.65035MaRDI QIDQ805153FDOQ805153

Authors: W. Homberg, Juergen-Fr. Hake 

Publication date: 1991

Published in: Parallel Computing (Search for Journal in Brave)

Recommendations

• Improved algorithm of matrix multiplication based on memory
• The combinatorics of cache misses during matrix multiplication
• scientific article; zbMATH DE number 823632
• A memetic algorithm to schedule planned maintenance for the national grid
• Influence of memory systems on vector processor performance
• OPTIMIZING THE MATRIX MULTIPLICATION PARALLEL ALGORITHMS ON A DISTRIBUTED-MEMORY MIMD MULTIPROCESSOR
• Effects of Ordering Strategies and Programming Paradigms on Sparse Matrix Computations

zbMATH Keywords

cacheeigenvalue problemslinear algebranumerical methodslinear equationsmatrix multiplicationmathematical softwarevector computerparallel computerIBMCRAYFUJITSUinterleaved memoryshortest-path problemstranslation lookaside buffer

Mathematics Subject Classification ID

Parallel numerical computation (65Y05) Information storage and retrieval of data (68P20) Numerical algorithms for specific classes of architectures (65Y10)

Cited In (2)

• Memory conflict resolution in vector supercomputers
• Modelling and measurement of memory access in Siemens VP supercomputers

Uses Software

• Algorithm 656

This page was built for publication: The impact of memory organization on the performance of matrix calculations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q805153)