Matches in SemOpenAlex for { <https://semopenalex.org/work/W4328005937> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4328005937 abstract "Edge computing is a promising computing paradigm that can reduce the burden of cloud servers and make it a powerful supplement to cloud computing. However, the application of edge computing is often limited because of the poor capability of the edge devices and high latency among them. And in practice, application requests are of a wide variety and often consist of sub-tasks with complex dependencies, making their characteristic difficult to learn. To improve the quality of service, it is necessary to balance computational overhead and latency. For those applications that are latency-sensitive, sometimes we need them to be completed before their deadlines. Considering that each sub-task within the request often has certain dependencies, we model the coming requests as co-tasks, which means that a request can be considered complete if and only if all sub-tasks within the request are complete. Our work aims to reduce the deadline-missing rate of the requests arriving at the cluster. In view of the complexity of this problem, and considering that the calling frequency of the scheduling algorithm is often high in practice, our work firstly applies the reinforcement learning method to configure the functions, after that we apply the greedy approach to dispatch the sub-tasks of each request to the machines. We call this reinforcement-learning-based algorithms Chaser. The experimental results show that our approach can reduce the deadline-missing rate by about 38%, 34% and 21% compared with the random and other two heuristic approaches in different situations." @default.
- W4328005937 created "2023-03-22" @default.
- W4328005937 creator A5031301468 @default.
- W4328005937 creator A5067465324 @default.
- W4328005937 date "2022-08-01" @default.
- W4328005937 modified "2023-09-27" @default.
- W4328005937 title "Online Learning-Based Co-Task Scheduling for Minimizing Deadline-Missing Rate in Edge Computing" @default.
- W4328005937 cites W1977785545 @default.
- W4328005937 cites W2035978754 @default.
- W4328005937 cites W2091565802 @default.
- W4328005937 cites W2149294210 @default.
- W4328005937 cites W2762456226 @default.
- W4328005937 cites W2766447205 @default.
- W4328005937 cites W2895830705 @default.
- W4328005937 cites W2897807975 @default.
- W4328005937 cites W2920088500 @default.
- W4328005937 cites W2920582503 @default.
- W4328005937 cites W2968986602 @default.
- W4328005937 cites W2982346115 @default.
- W4328005937 cites W3133317408 @default.
- W4328005937 doi "https://doi.org/10.1109/bigcom57025.2022.00016" @default.
- W4328005937 hasPublicationYear "2022" @default.
- W4328005937 type Work @default.
- W4328005937 citedByCount "0" @default.
- W4328005937 crossrefType "proceedings-article" @default.
- W4328005937 hasAuthorship W4328005937A5031301468 @default.
- W4328005937 hasAuthorship W4328005937A5067465324 @default.
- W4328005937 hasConcept C111919701 @default.
- W4328005937 hasConcept C120314980 @default.
- W4328005937 hasConcept C154945302 @default.
- W4328005937 hasConcept C162324750 @default.
- W4328005937 hasConcept C206729178 @default.
- W4328005937 hasConcept C21547014 @default.
- W4328005937 hasConcept C2778456923 @default.
- W4328005937 hasConcept C31258907 @default.
- W4328005937 hasConcept C41008148 @default.
- W4328005937 hasConcept C5119721 @default.
- W4328005937 hasConcept C76155785 @default.
- W4328005937 hasConcept C79974875 @default.
- W4328005937 hasConcept C82876162 @default.
- W4328005937 hasConcept C93996380 @default.
- W4328005937 hasConcept C97541855 @default.
- W4328005937 hasConceptScore W4328005937C111919701 @default.
- W4328005937 hasConceptScore W4328005937C120314980 @default.
- W4328005937 hasConceptScore W4328005937C154945302 @default.
- W4328005937 hasConceptScore W4328005937C162324750 @default.
- W4328005937 hasConceptScore W4328005937C206729178 @default.
- W4328005937 hasConceptScore W4328005937C21547014 @default.
- W4328005937 hasConceptScore W4328005937C2778456923 @default.
- W4328005937 hasConceptScore W4328005937C31258907 @default.
- W4328005937 hasConceptScore W4328005937C41008148 @default.
- W4328005937 hasConceptScore W4328005937C5119721 @default.
- W4328005937 hasConceptScore W4328005937C76155785 @default.
- W4328005937 hasConceptScore W4328005937C79974875 @default.
- W4328005937 hasConceptScore W4328005937C82876162 @default.
- W4328005937 hasConceptScore W4328005937C93996380 @default.
- W4328005937 hasConceptScore W4328005937C97541855 @default.
- W4328005937 hasLocation W43280059371 @default.
- W4328005937 hasOpenAccess W4328005937 @default.
- W4328005937 hasPrimaryLocation W43280059371 @default.
- W4328005937 hasRelatedWork W2889972190 @default.
- W4328005937 hasRelatedWork W2945616868 @default.
- W4328005937 hasRelatedWork W2970866953 @default.
- W4328005937 hasRelatedWork W2971330993 @default.
- W4328005937 hasRelatedWork W3175605422 @default.
- W4328005937 hasRelatedWork W4200573894 @default.
- W4328005937 hasRelatedWork W4210409506 @default.
- W4328005937 hasRelatedWork W4298234837 @default.
- W4328005937 hasRelatedWork W4312714354 @default.
- W4328005937 hasRelatedWork W4321844361 @default.
- W4328005937 isParatext "false" @default.
- W4328005937 isRetracted "false" @default.
- W4328005937 workType "article" @default.