Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378804782> ?p ?o ?g. }
Showing items 1 to 62 of
62
with 100 items per page.
- W4378804782 endingPage "1" @default.
- W4378804782 startingPage "1" @default.
- W4378804782 abstract "Sparse-matrix sparse-matrix multiplication (SpMM) is an important kernel in multiple areas, e.g., data analytics and machine learning. Due to the low on-chip memory requirement, the consistent data format, and the simplified control logic, the Gustavson’s algorithm is a promising backbone algorithm for SpMM on hardware accelerators. However, the off-chip memory traffic still limits the performance of the algorithm, especially on embedded FPGAs. Previous researchers optimize the Gustavson’s algorithm targeting high bandwidth memory-based architectures and their solutions cannot be directly applied to embedded FPGAs with traditional DDRs. In this work, we propose an efficient Gustavson-based sparse matrix-matrix multiplication accelerator on embedded FPGAs. The proposed design fully considers the feature of off-chip memory access on embedded FPGAs and the dataflow of the Gustavson’s algorithm. At first, we analyze the parallelism of the algorithm and propose to perform the algorithm with element-wise parallelism, which reduces the idle time of processing elements caused by synchronization. Further, we show a counter-intuitive example that the traditional cache leads to worse performance. Then, we propose a novel access pattern-aware cache scheme called SpCache, which provides quick responses to reduce bank conflicts caused by irregular memory accesses and combines streaming and caching to handle requests that access ordered elements of unpredictable length. Moreover, we propose to perform the merge on part of partial results, which removes some redundant merges in the naive implementation and has little postprocessing overhead. Finally, we conduct experiments on the Xilinx Zynq-UltraScale ZCU106 platform with a set of benchmarks from the SuiteSparse matrix collection. The experimental results show that the proposed design achieves an average 1.75x performance speedup compared to the baseline." @default.
- W4378804782 created "2023-06-01" @default.
- W4378804782 creator A5038598144 @default.
- W4378804782 creator A5063500802 @default.
- W4378804782 creator A5086240263 @default.
- W4378804782 date "2023-01-01" @default.
- W4378804782 modified "2023-09-30" @default.
- W4378804782 title "An Efficient Gustavson-based Sparse Matrix-matrix Multiplication Accelerator on Embedded FPGAs" @default.
- W4378804782 doi "https://doi.org/10.1109/tcad.2023.3281719" @default.
- W4378804782 hasPublicationYear "2023" @default.
- W4378804782 type Work @default.
- W4378804782 citedByCount "0" @default.
- W4378804782 crossrefType "journal-article" @default.
- W4378804782 hasAuthorship W4378804782A5038598144 @default.
- W4378804782 hasAuthorship W4378804782A5063500802 @default.
- W4378804782 hasAuthorship W4378804782A5086240263 @default.
- W4378804782 hasConcept C11413529 @default.
- W4378804782 hasConcept C115537543 @default.
- W4378804782 hasConcept C121332964 @default.
- W4378804782 hasConcept C149635348 @default.
- W4378804782 hasConcept C163716315 @default.
- W4378804782 hasConcept C17349429 @default.
- W4378804782 hasConcept C173608175 @default.
- W4378804782 hasConcept C41008148 @default.
- W4378804782 hasConcept C42935608 @default.
- W4378804782 hasConcept C56372850 @default.
- W4378804782 hasConcept C62520636 @default.
- W4378804782 hasConcept C84114770 @default.
- W4378804782 hasConcept C9390403 @default.
- W4378804782 hasConcept C96324660 @default.
- W4378804782 hasConceptScore W4378804782C11413529 @default.
- W4378804782 hasConceptScore W4378804782C115537543 @default.
- W4378804782 hasConceptScore W4378804782C121332964 @default.
- W4378804782 hasConceptScore W4378804782C149635348 @default.
- W4378804782 hasConceptScore W4378804782C163716315 @default.
- W4378804782 hasConceptScore W4378804782C17349429 @default.
- W4378804782 hasConceptScore W4378804782C173608175 @default.
- W4378804782 hasConceptScore W4378804782C41008148 @default.
- W4378804782 hasConceptScore W4378804782C42935608 @default.
- W4378804782 hasConceptScore W4378804782C56372850 @default.
- W4378804782 hasConceptScore W4378804782C62520636 @default.
- W4378804782 hasConceptScore W4378804782C84114770 @default.
- W4378804782 hasConceptScore W4378804782C9390403 @default.
- W4378804782 hasConceptScore W4378804782C96324660 @default.
- W4378804782 hasLocation W43788047821 @default.
- W4378804782 hasOpenAccess W4378804782 @default.
- W4378804782 hasPrimaryLocation W43788047821 @default.
- W4378804782 hasRelatedWork W1992476323 @default.
- W4378804782 hasRelatedWork W2040556424 @default.
- W4378804782 hasRelatedWork W2102248224 @default.
- W4378804782 hasRelatedWork W2142496304 @default.
- W4378804782 hasRelatedWork W2216820207 @default.
- W4378804782 hasRelatedWork W2547609661 @default.
- W4378804782 hasRelatedWork W2768573991 @default.
- W4378804782 hasRelatedWork W2899919874 @default.
- W4378804782 hasRelatedWork W3123222496 @default.
- W4378804782 hasRelatedWork W3131083367 @default.
- W4378804782 isParatext "false" @default.
- W4378804782 isRetracted "false" @default.
- W4378804782 workType "article" @default.