Exploiting parallelism in matrix-computation kernels for symmetric multiprocessor systems: Matrix-multiplication and matrix-addition algorithm optimizations by software pipelining and threads allocation.
Paolo D'AlbertoMarco BodratoAlexandru NicolauPublished in: ACM Trans. Math. Softw. (2011)
Keyphrases
- matrix multiplication
- distributed memory
- multiprocessor systems
- dynamic programming
- computational complexity
- message passing
- shared memory
- positive definite
- memory requirements
- parallel computation
- symmetric matrices
- matrix factorization
- parallel processing
- matching algorithm
- energy function
- kernel function
- probabilistic model