Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378942706> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4378942706 abstract "As the increasing complexity of Neural Network(NN) models leads to high demands for computation, AMD introduces a heterogeneous programmable system-on-chip (SoC), i.e., Versal ACAP architectures featured with programmable logic (PL), CPUs, and dedicated AI engines (AIE) ASICs which has a theoretical throughput up to 6.4 TFLOPs for FP32, 25.6 TOPs for INT16 and 102.4 TOPs for INT8. However, the higher level of complexity makes it non-trivial to achieve the theoretical performance even for well-studied applications like matrix-matrix multiply. In this paper, we provide AutoMM, an automatic white-box framework that can systematically generate the design for MM accelerators on Versal which achieves 3.7 TFLOPs, 7.5 TOPs, and 28.2 TOPs for FP32, INT16, and INT8 data type respectively. Our designs are tested on board and achieve gains of 7.20x (FP32), 3.26x (INT16), 6.23x (INT8) energy efficiency than AMD U250 FPGA, 2.32x (FP32) than Nvidia Jetson TX2 GPU, 1.06x (FP32), 1.70x (INT8) than Nvidia A100 GPU." @default.
- W4378942706 created "2023-06-01" @default.
- W4378942706 creator A5063866156 @default.
- W4378942706 creator A5074853634 @default.
- W4378942706 creator A5088193007 @default.
- W4378942706 date "2023-05-29" @default.
- W4378942706 modified "2023-10-16" @default.
- W4378942706 title "AutoMM: Energy-Efficient Multi-Data-Type Matrix Multiply Design on Heterogeneous Programmable System-on-Chip" @default.
- W4378942706 doi "https://doi.org/10.48550/arxiv.2305.18698" @default.
- W4378942706 hasPublicationYear "2023" @default.
- W4378942706 type Work @default.
- W4378942706 citedByCount "0" @default.
- W4378942706 crossrefType "posted-content" @default.
- W4378942706 hasAuthorship W4378942706A5063866156 @default.
- W4378942706 hasAuthorship W4378942706A5074853634 @default.
- W4378942706 hasAuthorship W4378942706A5088193007 @default.
- W4378942706 hasBestOaLocation W43789427061 @default.
- W4378942706 hasConcept C105795698 @default.
- W4378942706 hasConcept C106487976 @default.
- W4378942706 hasConcept C11413529 @default.
- W4378942706 hasConcept C118524514 @default.
- W4378942706 hasConcept C119599485 @default.
- W4378942706 hasConcept C127413603 @default.
- W4378942706 hasConcept C149635348 @default.
- W4378942706 hasConcept C154815118 @default.
- W4378942706 hasConcept C159985019 @default.
- W4378942706 hasConcept C165005293 @default.
- W4378942706 hasConcept C173608175 @default.
- W4378942706 hasConcept C186370098 @default.
- W4378942706 hasConcept C192562407 @default.
- W4378942706 hasConcept C2742236 @default.
- W4378942706 hasConcept C2777675136 @default.
- W4378942706 hasConcept C33923547 @default.
- W4378942706 hasConcept C41008148 @default.
- W4378942706 hasConcept C42935608 @default.
- W4378942706 hasConcept C45374587 @default.
- W4378942706 hasConcept C459310 @default.
- W4378942706 hasConcept C76155785 @default.
- W4378942706 hasConcept C77390884 @default.
- W4378942706 hasConcept C78519656 @default.
- W4378942706 hasConcept C9390403 @default.
- W4378942706 hasConceptScore W4378942706C105795698 @default.
- W4378942706 hasConceptScore W4378942706C106487976 @default.
- W4378942706 hasConceptScore W4378942706C11413529 @default.
- W4378942706 hasConceptScore W4378942706C118524514 @default.
- W4378942706 hasConceptScore W4378942706C119599485 @default.
- W4378942706 hasConceptScore W4378942706C127413603 @default.
- W4378942706 hasConceptScore W4378942706C149635348 @default.
- W4378942706 hasConceptScore W4378942706C154815118 @default.
- W4378942706 hasConceptScore W4378942706C159985019 @default.
- W4378942706 hasConceptScore W4378942706C165005293 @default.
- W4378942706 hasConceptScore W4378942706C173608175 @default.
- W4378942706 hasConceptScore W4378942706C186370098 @default.
- W4378942706 hasConceptScore W4378942706C192562407 @default.
- W4378942706 hasConceptScore W4378942706C2742236 @default.
- W4378942706 hasConceptScore W4378942706C2777675136 @default.
- W4378942706 hasConceptScore W4378942706C33923547 @default.
- W4378942706 hasConceptScore W4378942706C41008148 @default.
- W4378942706 hasConceptScore W4378942706C42935608 @default.
- W4378942706 hasConceptScore W4378942706C45374587 @default.
- W4378942706 hasConceptScore W4378942706C459310 @default.
- W4378942706 hasConceptScore W4378942706C76155785 @default.
- W4378942706 hasConceptScore W4378942706C77390884 @default.
- W4378942706 hasConceptScore W4378942706C78519656 @default.
- W4378942706 hasConceptScore W4378942706C9390403 @default.
- W4378942706 hasLocation W43789427061 @default.
- W4378942706 hasOpenAccess W4378942706 @default.
- W4378942706 hasPrimaryLocation W43789427061 @default.
- W4378942706 hasRelatedWork W1589309932 @default.
- W4378942706 hasRelatedWork W2100470915 @default.
- W4378942706 hasRelatedWork W2104020799 @default.
- W4378942706 hasRelatedWork W2114354660 @default.
- W4378942706 hasRelatedWork W2134422574 @default.
- W4378942706 hasRelatedWork W2966987601 @default.
- W4378942706 hasRelatedWork W2984236338 @default.
- W4378942706 hasRelatedWork W3010492628 @default.
- W4378942706 hasRelatedWork W1808933589 @default.
- W4378942706 hasRelatedWork W2506672464 @default.
- W4378942706 isParatext "false" @default.
- W4378942706 isRetracted "false" @default.
- W4378942706 workType "article" @default.