Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312996980> ?p ?o ?g. }
- W4312996980 abstract "Deep Learning Recommendation Models (DLRM) are widespread, account for a considerable data center footprint, and grow by more than 1.5x per year. With model size soon to be in terabytes range, leveraging Storage Class Memory (SCM) for inference enables lower power consumption. This paper evaluates the major challenges in extending the memory hierarchy to SCM for DLRM, and presents different techniques to improve performance through a Software Defined Memory. We show how underlying technologies such as Nand Flash and 3DXP differentiate, and relate to real world scenarios, enabling from 5% to 29% power savings." @default.
- W4312996980 created "2023-01-05" @default.
- W4312996980 creator A5007553043 @default.
- W4312996980 creator A5007892541 @default.
- W4312996980 creator A5008908132 @default.
- W4312996980 creator A5020495890 @default.
- W4312996980 creator A5024109615 @default.
- W4312996980 creator A5025118822 @default.
- W4312996980 creator A5038603364 @default.
- W4312996980 creator A5039508118 @default.
- W4312996980 creator A5046763803 @default.
- W4312996980 creator A5054053794 @default.
- W4312996980 creator A5060314107 @default.
- W4312996980 creator A5060639309 @default.
- W4312996980 creator A5066366329 @default.
- W4312996980 creator A5066494779 @default.
- W4312996980 creator A5068114879 @default.
- W4312996980 creator A5071949326 @default.
- W4312996980 creator A5086372658 @default.
- W4312996980 creator A5089808352 @default.
- W4312996980 creator A5090143435 @default.
- W4312996980 creator A5091185032 @default.
- W4312996980 date "2022-07-01" @default.
- W4312996980 modified "2023-10-03" @default.
- W4312996980 title "Supporting Massive DLRM Inference through Software Defined Memory" @default.
- W4312996980 cites W2210543184 @default.
- W4312996980 cites W2475334473 @default.
- W4312996980 cites W2512971201 @default.
- W4312996980 cites W2614794251 @default.
- W4312996980 cites W2794670651 @default.
- W4312996980 cites W2934584470 @default.
- W4312996980 cites W2973172293 @default.
- W4312996980 cites W2981587687 @default.
- W4312996980 cites W3016842236 @default.
- W4312996980 cites W3043590771 @default.
- W4312996980 cites W3155243801 @default.
- W4312996980 cites W3158783012 @default.
- W4312996980 cites W3169109617 @default.
- W4312996980 cites W3173720884 @default.
- W4312996980 cites W3188357540 @default.
- W4312996980 cites W3206341493 @default.
- W4312996980 doi "https://doi.org/10.1109/icdcs54860.2022.00037" @default.
- W4312996980 hasPublicationYear "2022" @default.
- W4312996980 type Work @default.
- W4312996980 citedByCount "3" @default.
- W4312996980 countsByYear W43129969802022 @default.
- W4312996980 countsByYear W43129969802023 @default.
- W4312996980 crossrefType "proceedings-article" @default.
- W4312996980 hasAuthorship W4312996980A5007553043 @default.
- W4312996980 hasAuthorship W4312996980A5007892541 @default.
- W4312996980 hasAuthorship W4312996980A5008908132 @default.
- W4312996980 hasAuthorship W4312996980A5020495890 @default.
- W4312996980 hasAuthorship W4312996980A5024109615 @default.
- W4312996980 hasAuthorship W4312996980A5025118822 @default.
- W4312996980 hasAuthorship W4312996980A5038603364 @default.
- W4312996980 hasAuthorship W4312996980A5039508118 @default.
- W4312996980 hasAuthorship W4312996980A5046763803 @default.
- W4312996980 hasAuthorship W4312996980A5054053794 @default.
- W4312996980 hasAuthorship W4312996980A5060314107 @default.
- W4312996980 hasAuthorship W4312996980A5060639309 @default.
- W4312996980 hasAuthorship W4312996980A5066366329 @default.
- W4312996980 hasAuthorship W4312996980A5066494779 @default.
- W4312996980 hasAuthorship W4312996980A5068114879 @default.
- W4312996980 hasAuthorship W4312996980A5071949326 @default.
- W4312996980 hasAuthorship W4312996980A5086372658 @default.
- W4312996980 hasAuthorship W4312996980A5089808352 @default.
- W4312996980 hasAuthorship W4312996980A5090143435 @default.
- W4312996980 hasAuthorship W4312996980A5091185032 @default.
- W4312996980 hasBestOaLocation W43129969802 @default.
- W4312996980 hasConcept C111919701 @default.
- W4312996980 hasConcept C115537543 @default.
- W4312996980 hasConcept C118524514 @default.
- W4312996980 hasConcept C132943942 @default.
- W4312996980 hasConcept C151730666 @default.
- W4312996980 hasConcept C154945302 @default.
- W4312996980 hasConcept C176649486 @default.
- W4312996980 hasConcept C199683683 @default.
- W4312996980 hasConcept C2776214188 @default.
- W4312996980 hasConcept C2777212361 @default.
- W4312996980 hasConcept C2777904410 @default.
- W4312996980 hasConcept C2778100165 @default.
- W4312996980 hasConcept C41008148 @default.
- W4312996980 hasConcept C74912251 @default.
- W4312996980 hasConcept C86803240 @default.
- W4312996980 hasConcept C98986596 @default.
- W4312996980 hasConceptScore W4312996980C111919701 @default.
- W4312996980 hasConceptScore W4312996980C115537543 @default.
- W4312996980 hasConceptScore W4312996980C118524514 @default.
- W4312996980 hasConceptScore W4312996980C132943942 @default.
- W4312996980 hasConceptScore W4312996980C151730666 @default.
- W4312996980 hasConceptScore W4312996980C154945302 @default.
- W4312996980 hasConceptScore W4312996980C176649486 @default.
- W4312996980 hasConceptScore W4312996980C199683683 @default.
- W4312996980 hasConceptScore W4312996980C2776214188 @default.
- W4312996980 hasConceptScore W4312996980C2777212361 @default.
- W4312996980 hasConceptScore W4312996980C2777904410 @default.
- W4312996980 hasConceptScore W4312996980C2778100165 @default.
- W4312996980 hasConceptScore W4312996980C41008148 @default.
- W4312996980 hasConceptScore W4312996980C74912251 @default.
- W4312996980 hasConceptScore W4312996980C86803240 @default.