{"title":"A CMP Thread Scheduling Strategy Based on Improved Firework Algorithm","authors":"Li Jingmei, L. Lanting, Wang Jiaxiang, Li He","doi":"10.1109/CIACT.2018.8480208","DOIUrl":null,"url":null,"abstract":"To fully exploit the parallelism and efficiency of CMP architecture, improve the efficiency of thread scheduling under CMP architecture and avoid the drop of multi-core processor throughput due to resource contention, a thread scheduling strategy - IFWAC based on improved fireworks algorithm is proposed. IFWAC excludes the fireworks with the optimal value of each generation to ensure the effectiveness of the sparks and avoid the waste of computing resources; The nonlinear inertia weighting factor is applied to the dimension of the rejection operation. So that the algorithm in the early iteration has a stronger global exploration capabilities. In the latter part of the iteration can be meticulous optimization. The IFWAC, the fireworks algorithm and the genetic algorithm are used to perform the scheduling experiment on the Matlab platform. The experimental results show that the IFWAC has obvious advantages in the convergence speed and the accuracy of the solution. The IFWAC improve the throughput of multi-core processors and has a wide range of research value.","PeriodicalId":358555,"journal":{"name":"2018 4th International Conference on Computational Intelligence & Communication Technology (CICT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 4th International Conference on Computational Intelligence & Communication Technology (CICT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CIACT.2018.8480208","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
To fully exploit the parallelism and efficiency of CMP architecture, improve the efficiency of thread scheduling under CMP architecture and avoid the drop of multi-core processor throughput due to resource contention, a thread scheduling strategy - IFWAC based on improved fireworks algorithm is proposed. IFWAC excludes the fireworks with the optimal value of each generation to ensure the effectiveness of the sparks and avoid the waste of computing resources; The nonlinear inertia weighting factor is applied to the dimension of the rejection operation. So that the algorithm in the early iteration has a stronger global exploration capabilities. In the latter part of the iteration can be meticulous optimization. The IFWAC, the fireworks algorithm and the genetic algorithm are used to perform the scheduling experiment on the Matlab platform. The experimental results show that the IFWAC has obvious advantages in the convergence speed and the accuracy of the solution. The IFWAC improve the throughput of multi-core processors and has a wide range of research value.