Matches in SemOpenAlex for { <https://semopenalex.org/work/W3026834348> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3026834348 abstract "Large-scale matrix multiplication is a fundamental kernel in science and engineering applications. However, existing computing platforms such as CPU, GPU and FPGA suffer from limited performance or excessive power consumption. This paper presents a high-performance and efficient accelerator named MMA for floating-point matrix multiplication based on scaled- out multi-array systolic arrays. A scheduling method is proposed for efficiently performing large-scale matrix-multiplication. Besides, an analytical model is built based on related design parameters to explore and determine the optimal design space. Evaluation results show that the accelerator with 8x8 matrix processing arrays nested with 8x8 systolic array can achieve a maximum performance of 12 TFLOPS and an efficiency of 99% for large-scale matrix multiplication. Compared with NVIDIA TESLA K4O GPU implemented with similar process, the actual performance of MMA is 2.57x of K4O, while the area is only about 58.4% of the latter." @default.
- W3026834348 created "2020-05-29" @default.
- W3026834348 creator A5016774224 @default.
- W3026834348 creator A5072960295 @default.
- W3026834348 creator A5077165659 @default.
- W3026834348 date "2019-10-01" @default.
- W3026834348 modified "2023-09-26" @default.
- W3026834348 title "A High-Performance Multi-array Accelerator for Large-Scale Floating-Point Matrix Multiplication" @default.
- W3026834348 cites W1582059966 @default.
- W3026834348 cites W1998203035 @default.
- W3026834348 cites W2014892800 @default.
- W3026834348 cites W2017369466 @default.
- W3026834348 cites W2061624656 @default.
- W3026834348 cites W2080275298 @default.
- W3026834348 cites W2110656410 @default.
- W3026834348 cites W2154394613 @default.
- W3026834348 cites W2165354374 @default.
- W3026834348 cites W2169631286 @default.
- W3026834348 cites W2963140246 @default.
- W3026834348 cites W4229992011 @default.
- W3026834348 cites W4249774071 @default.
- W3026834348 doi "https://doi.org/10.1109/eitce47263.2019.9095069" @default.
- W3026834348 hasPublicationYear "2019" @default.
- W3026834348 type Work @default.
- W3026834348 sameAs 3026834348 @default.
- W3026834348 citedByCount "0" @default.
- W3026834348 crossrefType "proceedings-article" @default.
- W3026834348 hasAuthorship W3026834348A5016774224 @default.
- W3026834348 hasAuthorship W3026834348A5072960295 @default.
- W3026834348 hasAuthorship W3026834348A5077165659 @default.
- W3026834348 hasConcept C106487976 @default.
- W3026834348 hasConcept C11413529 @default.
- W3026834348 hasConcept C114614502 @default.
- W3026834348 hasConcept C121332964 @default.
- W3026834348 hasConcept C126255220 @default.
- W3026834348 hasConcept C14580979 @default.
- W3026834348 hasConcept C149635348 @default.
- W3026834348 hasConcept C150741067 @default.
- W3026834348 hasConcept C159985019 @default.
- W3026834348 hasConcept C17349429 @default.
- W3026834348 hasConcept C173608175 @default.
- W3026834348 hasConcept C192562407 @default.
- W3026834348 hasConcept C206729178 @default.
- W3026834348 hasConcept C2780595030 @default.
- W3026834348 hasConcept C33923547 @default.
- W3026834348 hasConcept C35912277 @default.
- W3026834348 hasConcept C41008148 @default.
- W3026834348 hasConcept C42935608 @default.
- W3026834348 hasConcept C459310 @default.
- W3026834348 hasConcept C62520636 @default.
- W3026834348 hasConcept C74193536 @default.
- W3026834348 hasConcept C84114770 @default.
- W3026834348 hasConcept C84211073 @default.
- W3026834348 hasConcept C9390403 @default.
- W3026834348 hasConceptScore W3026834348C106487976 @default.
- W3026834348 hasConceptScore W3026834348C11413529 @default.
- W3026834348 hasConceptScore W3026834348C114614502 @default.
- W3026834348 hasConceptScore W3026834348C121332964 @default.
- W3026834348 hasConceptScore W3026834348C126255220 @default.
- W3026834348 hasConceptScore W3026834348C14580979 @default.
- W3026834348 hasConceptScore W3026834348C149635348 @default.
- W3026834348 hasConceptScore W3026834348C150741067 @default.
- W3026834348 hasConceptScore W3026834348C159985019 @default.
- W3026834348 hasConceptScore W3026834348C17349429 @default.
- W3026834348 hasConceptScore W3026834348C173608175 @default.
- W3026834348 hasConceptScore W3026834348C192562407 @default.
- W3026834348 hasConceptScore W3026834348C206729178 @default.
- W3026834348 hasConceptScore W3026834348C2780595030 @default.
- W3026834348 hasConceptScore W3026834348C33923547 @default.
- W3026834348 hasConceptScore W3026834348C35912277 @default.
- W3026834348 hasConceptScore W3026834348C41008148 @default.
- W3026834348 hasConceptScore W3026834348C42935608 @default.
- W3026834348 hasConceptScore W3026834348C459310 @default.
- W3026834348 hasConceptScore W3026834348C62520636 @default.
- W3026834348 hasConceptScore W3026834348C74193536 @default.
- W3026834348 hasConceptScore W3026834348C84114770 @default.
- W3026834348 hasConceptScore W3026834348C84211073 @default.
- W3026834348 hasConceptScore W3026834348C9390403 @default.
- W3026834348 hasLocation W30268343481 @default.
- W3026834348 hasOpenAccess W3026834348 @default.
- W3026834348 hasPrimaryLocation W30268343481 @default.
- W3026834348 hasRelatedWork W1865141138 @default.
- W3026834348 hasRelatedWork W2114393091 @default.
- W3026834348 hasRelatedWork W2122485368 @default.
- W3026834348 hasRelatedWork W2157456954 @default.
- W3026834348 hasRelatedWork W2391308973 @default.
- W3026834348 hasRelatedWork W2550374317 @default.
- W3026834348 hasRelatedWork W2791204867 @default.
- W3026834348 hasRelatedWork W2999652941 @default.
- W3026834348 hasRelatedWork W3011295035 @default.
- W3026834348 hasRelatedWork W85829498 @default.
- W3026834348 isParatext "false" @default.
- W3026834348 isRetracted "false" @default.
- W3026834348 magId "3026834348" @default.
- W3026834348 workType "article" @default.