FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3025347955> ?p ?o ?g. }

• W3025347955 endingPage "106002" @default.
• W3025347955 startingPage "106002" @default.
• W3025347955 abstract "Abstract Networking has become a well-known performance bottleneck for distributed machine learning (DML). Although lots of works have focused on accelerating the communication process of DML, they ignore the impact of the physical network on the DML performance. Concurrently, optical circuit switches (OCSes) are increasingly applied in data centers and clusters, which can fundamentally improve DML performance. It is worth noting that the non-negligible OCS reconfiguration delay makes OCS scheduling algorithms have a great impact on the upper application performance. However, existing OCS scheduling solutions are not suitable for DML jobs due to the iterative nature of DML jobs and their interleaving characteristics of communication and computation stages. Therefore, in this paper, we study the online multi-job scheduling for DML in OCS networks. Firstly, we propose heaviest-load-first (HLF), a heuristic algorithm for intra-job scheduling, which is based on the fact that the completion time of flows on the heaviest load port has a significant impact on the job completion time. Furthermore, we present Shortest Weighted Remaining Time First (SWRTF) algorithm for inter-job scheduling. In SWRTF, an available DML job is scheduled when the served job moves from communication stage to the computation stage, which significantly improves the circuit utilization. Based on large-scale simulations, we demonstrate HLF can significantly reduce the iteration communication time by up to 64.97% compared to the state-of-the-art circuit scheduler Sunflow. Besides, SWRTF can save up to 42.9%, 54.2%, 27.2% of Weighted-Job-Completion-Time (WJCT) compared to Shortest-Job-First, Baraat and Weighted-First inter-job scheduling algorithms, respectively." @default.
• W3025347955 created "2020-05-21" @default.
• W3025347955 creator A5016039620 @default.
• W3025347955 creator A5022284072 @default.
• W3025347955 creator A5024438027 @default.
• W3025347955 creator A5041291833 @default.
• W3025347955 creator A5053945147 @default.
• W3025347955 creator A5075454046 @default.
• W3025347955 date "2020-08-01" @default.
• W3025347955 modified "2023-09-24" @default.
• W3025347955 title "Online job scheduling for distributed machine learning in optical circuit switch networks" @default.
• W3025347955 cites W2057332538 @default.
• W3025347955 cites W2130531694 @default.
• W3025347955 cites W2131613942 @default.
• W3025347955 cites W2268702383 @default.
• W3025347955 cites W2530249205 @default.
• W3025347955 cites W2559673548 @default.
• W3025347955 cites W2742956305 @default.
• W3025347955 cites W2743429249 @default.
• W3025347955 cites W2789335636 @default.
• W3025347955 cites W2798515322 @default.
• W3025347955 cites W2884061259 @default.
• W3025347955 cites W2891391603 @default.
• W3025347955 cites W2897065005 @default.
• W3025347955 cites W2897089376 @default.
• W3025347955 cites W2905959927 @default.
• W3025347955 cites W2915752173 @default.
• W3025347955 cites W2919632853 @default.
• W3025347955 cites W2937088522 @default.
• W3025347955 cites W2963390885 @default.
• W3025347955 cites W2972087877 @default.
• W3025347955 cites W2975712713 @default.
• W3025347955 cites W2978015420 @default.
• W3025347955 cites W3000032064 @default.
• W3025347955 cites W3009983062 @default.
• W3025347955 cites W3018555825 @default.
• W3025347955 cites W3047584849 @default.
• W3025347955 cites W4231891315 @default.
• W3025347955 cites W4235670058 @default.
• W3025347955 cites W4245826835 @default.
• W3025347955 cites W4251897646 @default.
• W3025347955 cites W4255119735 @default.
• W3025347955 cites W4289401659 @default.
• W3025347955 doi "https://doi.org/10.1016/j.knosys.2020.106002" @default.
• W3025347955 hasPublicationYear "2020" @default.
• W3025347955 type Work @default.
• W3025347955 sameAs 3025347955 @default.
• W3025347955 citedByCount "4" @default.
• W3025347955 countsByYear W30253479552022 @default.
• W3025347955 crossrefType "journal-article" @default.
• W3025347955 hasAuthorship W3025347955A5016039620 @default.
• W3025347955 hasAuthorship W3025347955A5022284072 @default.
• W3025347955 hasAuthorship W3025347955A5024438027 @default.
• W3025347955 hasAuthorship W3025347955A5041291833 @default.
• W3025347955 hasAuthorship W3025347955A5053945147 @default.
• W3025347955 hasAuthorship W3025347955A5075454046 @default.
• W3025347955 hasConcept C111873713 @default.
• W3025347955 hasConcept C120314980 @default.
• W3025347955 hasConcept C126255220 @default.
• W3025347955 hasConcept C160403385 @default.
• W3025347955 hasConcept C206729178 @default.
• W3025347955 hasConcept C31258907 @default.
• W3025347955 hasConcept C33923547 @default.
• W3025347955 hasConcept C41008148 @default.
• W3025347955 hasConceptScore W3025347955C111873713 @default.
• W3025347955 hasConceptScore W3025347955C120314980 @default.
• W3025347955 hasConceptScore W3025347955C126255220 @default.
• W3025347955 hasConceptScore W3025347955C160403385 @default.
• W3025347955 hasConceptScore W3025347955C206729178 @default.
• W3025347955 hasConceptScore W3025347955C31258907 @default.
• W3025347955 hasConceptScore W3025347955C33923547 @default.
• W3025347955 hasConceptScore W3025347955C41008148 @default.
• W3025347955 hasFunder F4320335777 @default.
• W3025347955 hasLocation W30253479551 @default.
• W3025347955 hasOpenAccess W3025347955 @default.
• W3025347955 hasPrimaryLocation W30253479551 @default.
• W3025347955 hasRelatedWork W1572108542 @default.
• W3025347955 hasRelatedWork W1882733036 @default.
• W3025347955 hasRelatedWork W2019385245 @default.
• W3025347955 hasRelatedWork W2039968861 @default.
• W3025347955 hasRelatedWork W2109998134 @default.
• W3025347955 hasRelatedWork W2125095596 @default.
• W3025347955 hasRelatedWork W2140625810 @default.
• W3025347955 hasRelatedWork W2160425906 @default.
• W3025347955 hasRelatedWork W2389719923 @default.
• W3025347955 hasRelatedWork W2546696010 @default.
• W3025347955 hasVolume "201-202" @default.
• W3025347955 isParatext "false" @default.
• W3025347955 isRetracted "false" @default.
• W3025347955 magId "3025347955" @default.
• W3025347955 workType "article" @default.