Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Jungwoo | - |
dc.contributor.author | Oh, Seonggyun | - |
dc.contributor.author | Kung, Jaeha | - |
dc.contributor.author | Kim, Yeseong | - |
dc.contributor.author | Lee, Sungjin | - |
dc.date.accessioned | 2024-07-19T14:40:12Z | - |
dc.date.available | 2024-07-19T14:40:12Z | - |
dc.date.created | 2024-05-17 | - |
dc.date.issued | 2024-05-01 | - |
dc.identifier.isbn | 9798400703867 | - |
dc.identifier.uri | http://hdl.handle.net/20.500.11750/56705 | - |
dc.description.abstract | This paper proposes a novel photo storage system called NDPipe, which accelerates the performance of training and inference for image data by leveraging near-data processing in photo storage servers. NDPipe distributes storage servers with inexpensive commodity GPUs in a data center and uses their collective intelligence to perform inference and training near image data. By efficiently partitioning deep neural network (DNN) models and exploiting the data parallelism of many storage servers, NDPipe can achieve high training throughput with low synchronization costs. NDPipe optimizes the near-data processing engine to maximally utilize system components in each storage server. Our results show that, given the same energy budget, NDPipe exhibits 1.39× higher inference throughput and 2.64× faster training speed than typical photo storage systems. © 2024 Copyright held by the owner/author(s). | - |
dc.language | English | - |
dc.publisher | Association for Computing Machinery | - |
dc.relation.ispartof | ASPLOS '24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3 | - |
dc.title | NDPipe: Exploiting Near-data Processing for Scalable Inference and Continuous Training in Photo Storage | - |
dc.type | Conference Paper | - |
dc.identifier.doi | 10.1145/3620666.3651345 | - |
dc.identifier.scopusid | 2-s2.0-85192147297 | - |
dc.identifier.bibliographicCitation | Architectural Support for Programming Languages and Operating Systems, pp.689 - 707 | - |
dc.identifier.url | https://www.asplos-conference.org/asplos2024/main-program/index.html#:~:text=.%20Lightning%20Talk-,NDPipe,-%3A%20Exploiting%20Near%2Ddata | - |
dc.citation.conferenceDate | 2024-04-27 | - |
dc.citation.conferencePlace | US | - |
dc.citation.conferencePlace | San Diego | - |
dc.citation.endPage | 707 | - |
dc.citation.startPage | 689 | - |
dc.citation.title | Architectural Support for Programming Languages and Operating Systems | - |