Matches in SemOpenAlex for { <https://semopenalex.org/work/W1508286210> ?p ?o ?g. }
- W1508286210 abstract "Compressed sparse row (CSR) is a frequently used format for sparse matrix storage. However, the state-of-the-art CSR-based sparse matrix-vector multiplication (SpMV) implementations on CUDA-enabled GPUs do not exhibit very high efficiency. This has motivated the development of some alternative storage formats for GPU computing. Unfortunately, these alternatives are incompatible with most CPU-centric programs and require dynamic conversion from CSR at runtime, thus incurring significant computational and storage overheads. We present LightSpMV, a novel CUDA-compatible SpMV algorithm using the standard CSR format, which achieves high speed by benefiting from the fine-grained dynamic distribution of matrix rows over warps/vectors. In LightSpMV, two dynamic row distribution approaches have been investigated at the vector and warp levels with atomic operations and warp shuffle functions as the fundamental building blocks. We have evaluated LightSpMV using various sparse matrices and further compared it to the CSR-based SpMV subprograms in the state-of-the-art CUSP and cuSPARSE libraries. Performance evaluation reveals that on the same Tesla K40c GPU, LightSpMV is superior to both CUSP and cuSPARSE, with a speedup of up to 2.60 and 2.63 over CUSP, and up to 1.93 and 1.79 over cuSPARSE for single and double precision, respectively. LightSpMV is available at http://lightspmv.sourceforge.net." @default.
- W1508286210 created "2016-06-24" @default.
- W1508286210 creator A5009074414 @default.
- W1508286210 creator A5020388832 @default.
- W1508286210 date "2015-07-01" @default.
- W1508286210 modified "2023-10-01" @default.
- W1508286210 title "LightSpMV: Faster CSR-based sparse matrix-vector multiplication on CUDA-enabled GPUs" @default.
- W1508286210 cites W1506342804 @default.
- W1508286210 cites W1515144947 @default.
- W1508286210 cites W1568272005 @default.
- W1508286210 cites W1768849904 @default.
- W1508286210 cites W2023930909 @default.
- W1508286210 cites W2031460602 @default.
- W1508286210 cites W2087507944 @default.
- W1508286210 cites W2088866486 @default.
- W1508286210 cites W2095836023 @default.
- W1508286210 cites W2128539477 @default.
- W1508286210 cites W2128853364 @default.
- W1508286210 cites W3141650078 @default.
- W1508286210 doi "https://doi.org/10.1109/asap.2015.7245713" @default.
- W1508286210 hasPublicationYear "2015" @default.
- W1508286210 type Work @default.
- W1508286210 sameAs 1508286210 @default.
- W1508286210 citedByCount "32" @default.
- W1508286210 countsByYear W15082862102015 @default.
- W1508286210 countsByYear W15082862102016 @default.
- W1508286210 countsByYear W15082862102017 @default.
- W1508286210 countsByYear W15082862102018 @default.
- W1508286210 countsByYear W15082862102019 @default.
- W1508286210 countsByYear W15082862102020 @default.
- W1508286210 countsByYear W15082862102021 @default.
- W1508286210 countsByYear W15082862102022 @default.
- W1508286210 countsByYear W15082862102023 @default.
- W1508286210 crossrefType "proceedings-article" @default.
- W1508286210 hasAuthorship W1508286210A5009074414 @default.
- W1508286210 hasAuthorship W1508286210A5020388832 @default.
- W1508286210 hasConcept C106487976 @default.
- W1508286210 hasConcept C11413529 @default.
- W1508286210 hasConcept C114614502 @default.
- W1508286210 hasConcept C121332964 @default.
- W1508286210 hasConcept C121684516 @default.
- W1508286210 hasConcept C159985019 @default.
- W1508286210 hasConcept C163716315 @default.
- W1508286210 hasConcept C17349429 @default.
- W1508286210 hasConcept C173608175 @default.
- W1508286210 hasConcept C192562407 @default.
- W1508286210 hasConcept C21442007 @default.
- W1508286210 hasConcept C2778119891 @default.
- W1508286210 hasConcept C2780595030 @default.
- W1508286210 hasConcept C33923547 @default.
- W1508286210 hasConcept C35912277 @default.
- W1508286210 hasConcept C41008148 @default.
- W1508286210 hasConcept C459310 @default.
- W1508286210 hasConcept C50630238 @default.
- W1508286210 hasConcept C56372850 @default.
- W1508286210 hasConcept C62520636 @default.
- W1508286210 hasConcept C68339613 @default.
- W1508286210 hasConcept C74193536 @default.
- W1508286210 hasConcept C84114770 @default.
- W1508286210 hasConcept C84211073 @default.
- W1508286210 hasConceptScore W1508286210C106487976 @default.
- W1508286210 hasConceptScore W1508286210C11413529 @default.
- W1508286210 hasConceptScore W1508286210C114614502 @default.
- W1508286210 hasConceptScore W1508286210C121332964 @default.
- W1508286210 hasConceptScore W1508286210C121684516 @default.
- W1508286210 hasConceptScore W1508286210C159985019 @default.
- W1508286210 hasConceptScore W1508286210C163716315 @default.
- W1508286210 hasConceptScore W1508286210C17349429 @default.
- W1508286210 hasConceptScore W1508286210C173608175 @default.
- W1508286210 hasConceptScore W1508286210C192562407 @default.
- W1508286210 hasConceptScore W1508286210C21442007 @default.
- W1508286210 hasConceptScore W1508286210C2778119891 @default.
- W1508286210 hasConceptScore W1508286210C2780595030 @default.
- W1508286210 hasConceptScore W1508286210C33923547 @default.
- W1508286210 hasConceptScore W1508286210C35912277 @default.
- W1508286210 hasConceptScore W1508286210C41008148 @default.
- W1508286210 hasConceptScore W1508286210C459310 @default.
- W1508286210 hasConceptScore W1508286210C50630238 @default.
- W1508286210 hasConceptScore W1508286210C56372850 @default.
- W1508286210 hasConceptScore W1508286210C62520636 @default.
- W1508286210 hasConceptScore W1508286210C68339613 @default.
- W1508286210 hasConceptScore W1508286210C74193536 @default.
- W1508286210 hasConceptScore W1508286210C84114770 @default.
- W1508286210 hasConceptScore W1508286210C84211073 @default.
- W1508286210 hasLocation W15082862101 @default.
- W1508286210 hasOpenAccess W1508286210 @default.
- W1508286210 hasPrimaryLocation W15082862101 @default.
- W1508286210 hasRelatedWork W1481877323 @default.
- W1508286210 hasRelatedWork W1501847821 @default.
- W1508286210 hasRelatedWork W1978647314 @default.
- W1508286210 hasRelatedWork W1981557297 @default.
- W1508286210 hasRelatedWork W2011159963 @default.
- W1508286210 hasRelatedWork W2186439059 @default.
- W1508286210 hasRelatedWork W2256144436 @default.
- W1508286210 hasRelatedWork W2791204867 @default.
- W1508286210 hasRelatedWork W3038415719 @default.
- W1508286210 hasRelatedWork W52302056 @default.
- W1508286210 isParatext "false" @default.
- W1508286210 isRetracted "false" @default.
- W1508286210 magId "1508286210" @default.