Optimizing the parallel computation of linear recurrences using compact matrix representations

Nistor, Adrian; Chin, Wei-Ngan; Tan, Tony; Țăpuș, Nicolae

doi:10.1016/j.jpdc.2009.01.004

Cited by 2 publications

(1 citation statement)

References 15 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nistor et al [16] presented optimization techniques for matrix multiplication derived from parallelized linear recurrence equations. Their techniques reduce the space used in matrix multiplication by recovering an expression of one entry from an expression of another entry on the basis of linear relations between these expressions.…”

Section: Redon and Feauriermentioning

confidence: 99%

Automatic parallelization via matrix multiplication

Sato¹,

Iwasaki²

2011

Proceedings of the 32nd ACM SIGPLAN Conference on Programming Language Design and Implementation

View full text Add to dashboard Cite

Existing work that deals with parallelization of complicated reductions and scans focuses only on formalism and hardly dealt with implementation. To bridge the gap between formalism and implementation, we have integrated parallelization via matrix multiplication into compiler construction. Our framework can deal with complicated loops that existing techniques in compilers cannot parallelize. Moreover, we have sophisticated our framework by developing two sets of techniques. One enhances its capability for parallelization by extracting max-operators automatically, and the other improves the performance of parallelized programs by eliminating redundancy. We have also implemented our framework and techniques as a parallelizer in a compiler. Experiments on examples that existing compilers cannot parallelize have demonstrated the scalability of programs parallelized by our implementation.

show abstract

Section: Redon and Feauriermentioning

confidence: 99%