FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3138708228> ?p ?o ?g. }

• W3138708228 endingPage "262" @default.
• W3138708228 startingPage "251" @default.
• W3138708228 abstract "The Deep Neural Network (DNN), Recurrent Neural Network (RNN) applications, rapidly becoming attractive to the market, process a large amount of low-locality data; thus, the memory bandwidth limits their peak performance. Therefore, many data centers actively adapt high-bandwidth memory like HBM2/HBM2E to resolve the problem. However, this approach would not provide a complete solution since it still transfers the data from the memory to the computing unit. Thus, processing-in-memory (PIM), which performs the computation inside memory, has attracted attention. However, most previous methods require the modification or the extension of core pipelines and memory system components like memory controllers, making the practical implementation of PIM very challenging and expensive in development. In this article, we propose a Silent-PIM that performs the PIM computation with standard DRAM memory requests; thus, requiring no hardware modifications and allowing the PIM memory device to perform the computation while servicing non-PIM applications' memory requests. We can achieve our design goal by preserving the standard memory request behaviors and satisfying the DRAM standard timing requirements. In addition, using standard memory requests makes it possible to use DMA as a PIM's offloading engine, resulting in processing the PIM memory requests fast and making a core perform other tasks. We compared the performance of three Long Short-Term Memory models (LSTM) kernels on real platforms, such as the Silent-PIM modeled on the FPGA, GPU, and CPU. For (p ×512) ×(512 ×2048) matrix multiplication with a batch size p varying from 1 to 128, the Silent-PIM performed up to 16.9x and 24.6x faster than GPU and CPU, respectively, p=1, which was the case without having any data reuse. At p=128, the highest data reuse case, the GPU performance was the highest, but the PIM performance was still higher than the CPU execution. Similarly, at (p ×2048) element-wise multiplication and addition, where there was no data reuse, the Silent-PIM always achieved higher than both CPU and GPU. It also showed that when the PIM's EDP performance was superior to the others in all the cases having no data reuse." @default.
• W3138708228 created "2021-03-29" @default.
• W3138708228 creator A5001628148 @default.
• W3138708228 creator A5005365762 @default.
• W3138708228 creator A5009921737 @default.
• W3138708228 creator A5051989657 @default.
• W3138708228 creator A5067977998 @default.
• W3138708228 creator A5078505862 @default.
• W3138708228 creator A5084900880 @default.
• W3138708228 date "2022-02-01" @default.
• W3138708228 modified "2023-09-26" @default.
• W3138708228 title "Silent-PIM: Realizing the Processing-in-Memory Computing With Standard Memory Requests" @default.
• W3138708228 cites W2014977566 @default.
• W3138708228 cites W2044788775 @default.
• W3138708228 cites W2059198010 @default.
• W3138708228 cites W2064675550 @default.
• W3138708228 cites W2086112773 @default.
• W3138708228 cites W2093524602 @default.
• W3138708228 cites W2155070484 @default.
• W3138708228 cites W2155385791 @default.
• W3138708228 cites W2162651880 @default.
• W3138708228 cites W2536999580 @default.
• W3138708228 cites W2612654866 @default.
• W3138708228 cites W26556108 @default.
• W3138708228 cites W2766489088 @default.
• W3138708228 cites W2776052384 @default.
• W3138708228 cites W2794243109 @default.
• W3138708228 cites W2799011136 @default.
• W3138708228 cites W2809205380 @default.
• W3138708228 cites W2811080765 @default.
• W3138708228 cites W2896090304 @default.
• W3138708228 cites W2904295992 @default.
• W3138708228 cites W2949792175 @default.
• W3138708228 cites W2997464342 @default.
• W3138708228 cites W3042598257 @default.
• W3138708228 doi "https://doi.org/10.1109/tpds.2021.3065365" @default.
• W3138708228 hasPublicationYear "2022" @default.
• W3138708228 type Work @default.
• W3138708228 sameAs 3138708228 @default.
• W3138708228 citedByCount "7" @default.
• W3138708228 countsByYear W31387082282021 @default.
• W3138708228 countsByYear W31387082282022 @default.
• W3138708228 countsByYear W31387082282023 @default.
• W3138708228 crossrefType "journal-article" @default.
• W3138708228 hasAuthorship W3138708228A5001628148 @default.
• W3138708228 hasAuthorship W3138708228A5005365762 @default.
• W3138708228 hasAuthorship W3138708228A5009921737 @default.
• W3138708228 hasAuthorship W3138708228A5051989657 @default.
• W3138708228 hasAuthorship W3138708228A5067977998 @default.
• W3138708228 hasAuthorship W3138708228A5078505862 @default.
• W3138708228 hasAuthorship W3138708228A5084900880 @default.
• W3138708228 hasConcept C118524514 @default.
• W3138708228 hasConcept C149635348 @default.
• W3138708228 hasConcept C152890283 @default.
• W3138708228 hasConcept C171675096 @default.
• W3138708228 hasConcept C173608175 @default.
• W3138708228 hasConcept C176649486 @default.
• W3138708228 hasConcept C188045654 @default.
• W3138708228 hasConcept C41008148 @default.
• W3138708228 hasConcept C57863822 @default.
• W3138708228 hasConcept C63511323 @default.
• W3138708228 hasConcept C74426580 @default.
• W3138708228 hasConcept C82687282 @default.
• W3138708228 hasConcept C87907426 @default.
• W3138708228 hasConcept C92855701 @default.
• W3138708228 hasConcept C93446704 @default.
• W3138708228 hasConcept C9390403 @default.
• W3138708228 hasConcept C98986596 @default.
• W3138708228 hasConceptScore W3138708228C118524514 @default.
• W3138708228 hasConceptScore W3138708228C149635348 @default.
• W3138708228 hasConceptScore W3138708228C152890283 @default.
• W3138708228 hasConceptScore W3138708228C171675096 @default.
• W3138708228 hasConceptScore W3138708228C173608175 @default.
• W3138708228 hasConceptScore W3138708228C176649486 @default.
• W3138708228 hasConceptScore W3138708228C188045654 @default.
• W3138708228 hasConceptScore W3138708228C41008148 @default.
• W3138708228 hasConceptScore W3138708228C57863822 @default.
• W3138708228 hasConceptScore W3138708228C63511323 @default.
• W3138708228 hasConceptScore W3138708228C74426580 @default.
• W3138708228 hasConceptScore W3138708228C82687282 @default.
• W3138708228 hasConceptScore W3138708228C87907426 @default.
• W3138708228 hasConceptScore W3138708228C92855701 @default.
• W3138708228 hasConceptScore W3138708228C93446704 @default.
• W3138708228 hasConceptScore W3138708228C9390403 @default.
• W3138708228 hasConceptScore W3138708228C98986596 @default.
• W3138708228 hasFunder F4320322064 @default.
• W3138708228 hasIssue "2" @default.
• W3138708228 hasLocation W31387082281 @default.
• W3138708228 hasOpenAccess W3138708228 @default.
• W3138708228 hasPrimaryLocation W31387082281 @default.
• W3138708228 hasRelatedWork W1575240748 @default.
• W3138708228 hasRelatedWork W1608814317 @default.
• W3138708228 hasRelatedWork W1837030695 @default.
• W3138708228 hasRelatedWork W2032235477 @default.
• W3138708228 hasRelatedWork W2084310805 @default.
• W3138708228 hasRelatedWork W2138825797 @default.
• W3138708228 hasRelatedWork W2491097902 @default.
• W3138708228 hasRelatedWork W2782503170 @default.

• next