{"title":"Resizing of Heterogeneous Platforms and the Optimization of Parallel Applications","authors":"Moussa Beji, Sami Achour","doi":"10.1109/PDP2018.2018.00029","DOIUrl":null,"url":null,"abstract":"With the birth of multi-cluster platforms, scheduling and finding the optimal number of resources (clusters, processors) to execute an application constitute very critical problems. In this paper, we address the need for scheduling techniques for parallel task applications on this kind of platforms and we propose a new strategy for scheduling sequential task graphs based on existing heuristics that have proved to be efficient on homogeneous environments. The contribution of this paper lies in determining the appropriate clusters which participate to compute a given application. Our solution is composed of three steps: Firstly, determining of the computing clusters, secondly, determining the optimal number of processors in each cluster, finally place the tasks on the appropriate processors. Simulation results, based on both randomly generated graphs and real configuration platforms, show that the proposed approach provides interesting trade-off between makespan and resource consumption.","PeriodicalId":333367,"journal":{"name":"2018 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2018-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PDP2018.2018.00029","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
With the birth of multi-cluster platforms, scheduling and finding the optimal number of resources (clusters, processors) to execute an application constitute very critical problems. In this paper, we address the need for scheduling techniques for parallel task applications on this kind of platforms and we propose a new strategy for scheduling sequential task graphs based on existing heuristics that have proved to be efficient on homogeneous environments. The contribution of this paper lies in determining the appropriate clusters which participate to compute a given application. Our solution is composed of three steps: Firstly, determining of the computing clusters, secondly, determining the optimal number of processors in each cluster, finally place the tasks on the appropriate processors. Simulation results, based on both randomly generated graphs and real configuration platforms, show that the proposed approach provides interesting trade-off between makespan and resource consumption.