Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897065005> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2897065005 abstract "In recent years, many companies have developed various distributed computation frameworks for processing machine learning (ML) jobs in clusters. Networking is a well-known bottleneck for ML systems and the cluster demands efficient scheduling for huge traffic (up to 1GB per flow) generated by ML jobs. Coflow has been proven an effective abstraction to schedule flows of such data-parallel applications. However, the implementation of coflow scheduling policy is constrained when coflow characteristics are unknown a prior, and when TCP congestion control misinterprets the congestion signal leading to low throughput. Fortunately, traffic patterns experienced by some ML jobs support to speculate the complete coflow characteristic with limited information. Hence this paper summarizes coflow from these ML jobs as self-similar coflow and proposes a decentralized self-similar coflow scheduler Cicada. Cicada assigns each coflow a probe flow to speculate its characteristics during the transportation and employs the Shortest Job First (SJF) to separate coflow into strict priority queues based on the speculation result. To achieve full bandwidth for throughput- sensitive ML jobs, and to guarantee the scheduling policy implementation, Cicada promotes the elastic transport-layer rate control that outperforms prior works. Large-scale simulations show that Cicada completes coflow 2.08x faster than the state-of-the-art schemes in the information-agnostic scenario." @default.
- W2897065005 created "2018-10-26" @default.
- W2897065005 creator A5033421780 @default.
- W2897065005 creator A5054845805 @default.
- W2897065005 creator A5059211235 @default.
- W2897065005 creator A5066046506 @default.
- W2897065005 creator A5084370010 @default.
- W2897065005 date "2018-07-01" @default.
- W2897065005 modified "2023-10-03" @default.
- W2897065005 title "Cross-Layer Self-Similar Coflow Scheduling for Machine Learning Clusters" @default.
- W2897065005 cites W1442374986 @default.
- W2897065005 cites W1464363888 @default.
- W2897065005 cites W1515106148 @default.
- W2897065005 cites W1522584278 @default.
- W2897065005 cites W1657185548 @default.
- W2897065005 cites W1746258828 @default.
- W2897065005 cites W1988150362 @default.
- W2897065005 cites W1999195717 @default.
- W2897065005 cites W2013932952 @default.
- W2897065005 cites W2056394733 @default.
- W2897065005 cites W2074988668 @default.
- W2897065005 cites W2083842231 @default.
- W2897065005 cites W2091520147 @default.
- W2897065005 cites W2108598243 @default.
- W2897065005 cites W2117884704 @default.
- W2897065005 cites W2141992894 @default.
- W2897065005 cites W2143021454 @default.
- W2897065005 cites W2149939304 @default.
- W2897065005 cites W2157990152 @default.
- W2897065005 cites W2164740236 @default.
- W2897065005 cites W2168427242 @default.
- W2897065005 cites W2186615578 @default.
- W2897065005 cites W2488801318 @default.
- W2897065005 cites W2504213755 @default.
- W2897065005 cites W2505830093 @default.
- W2897065005 cites W2540285146 @default.
- W2897065005 cites W2563188592 @default.
- W2897065005 cites W2589642470 @default.
- W2897065005 cites W2744628735 @default.
- W2897065005 cites W2964043796 @default.
- W2897065005 doi "https://doi.org/10.1109/icccn.2018.8487329" @default.
- W2897065005 hasPublicationYear "2018" @default.
- W2897065005 type Work @default.
- W2897065005 sameAs 2897065005 @default.
- W2897065005 citedByCount "2" @default.
- W2897065005 countsByYear W28970650052020 @default.
- W2897065005 countsByYear W28970650052022 @default.
- W2897065005 crossrefType "proceedings-article" @default.
- W2897065005 hasAuthorship W2897065005A5033421780 @default.
- W2897065005 hasAuthorship W2897065005A5054845805 @default.
- W2897065005 hasAuthorship W2897065005A5059211235 @default.
- W2897065005 hasAuthorship W2897065005A5066046506 @default.
- W2897065005 hasAuthorship W2897065005A5084370010 @default.
- W2897065005 hasConcept C111919701 @default.
- W2897065005 hasConcept C11413529 @default.
- W2897065005 hasConcept C120314980 @default.
- W2897065005 hasConcept C149635348 @default.
- W2897065005 hasConcept C160403385 @default.
- W2897065005 hasConcept C162324750 @default.
- W2897065005 hasConcept C206729178 @default.
- W2897065005 hasConcept C21547014 @default.
- W2897065005 hasConcept C2780513914 @default.
- W2897065005 hasConcept C31258907 @default.
- W2897065005 hasConcept C41008148 @default.
- W2897065005 hasConcept C45374587 @default.
- W2897065005 hasConcept C68387754 @default.
- W2897065005 hasConceptScore W2897065005C111919701 @default.
- W2897065005 hasConceptScore W2897065005C11413529 @default.
- W2897065005 hasConceptScore W2897065005C120314980 @default.
- W2897065005 hasConceptScore W2897065005C149635348 @default.
- W2897065005 hasConceptScore W2897065005C160403385 @default.
- W2897065005 hasConceptScore W2897065005C162324750 @default.
- W2897065005 hasConceptScore W2897065005C206729178 @default.
- W2897065005 hasConceptScore W2897065005C21547014 @default.
- W2897065005 hasConceptScore W2897065005C2780513914 @default.
- W2897065005 hasConceptScore W2897065005C31258907 @default.
- W2897065005 hasConceptScore W2897065005C41008148 @default.
- W2897065005 hasConceptScore W2897065005C45374587 @default.
- W2897065005 hasConceptScore W2897065005C68387754 @default.
- W2897065005 hasLocation W28970650051 @default.
- W2897065005 hasOpenAccess W2897065005 @default.
- W2897065005 hasPrimaryLocation W28970650051 @default.
- W2897065005 hasRelatedWork W1594290885 @default.
- W2897065005 hasRelatedWork W1594822307 @default.
- W2897065005 hasRelatedWork W1810925108 @default.
- W2897065005 hasRelatedWork W1998323989 @default.
- W2897065005 hasRelatedWork W2010361322 @default.
- W2897065005 hasRelatedWork W2109998134 @default.
- W2897065005 hasRelatedWork W2124870959 @default.
- W2897065005 hasRelatedWork W4230527247 @default.
- W2897065005 hasRelatedWork W2583476349 @default.
- W2897065005 hasRelatedWork W2805806777 @default.
- W2897065005 isParatext "false" @default.
- W2897065005 isRetracted "false" @default.
- W2897065005 magId "2897065005" @default.
- W2897065005 workType "article" @default.