{"title":"Simulation & Analysis of a Mean Response Time Upper-bound for Homogeneous Fork/Join Queues","authors":"R. Chen, K. Reschke, Muchenxuan Tong","doi":"10.1109/UKSIM.2011.88","DOIUrl":null,"url":null,"abstract":"In this paper, we study general K-queue first-in-first-out homogeneous fork/join queueing (HFJ) systems for any K ≥ 2. We simulate and analyze an upper-bound for the mean response time that we denote by T[K]. The upper-bound uses a relatively tiny-scale system to predict the performance of a huge-scale system. It is evaluated for 10-million queues on a regular HP-PC with Intel i7-860 for three different HFJ cases. The maximum time is 16 minutes, which is only about 0.01% of the full system simulation time. We show that it is fast, close, economical and consistent by comparison and analysis.","PeriodicalId":161995,"journal":{"name":"2011 UkSim 13th International Conference on Computer Modelling and Simulation","volume":"70 12","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2011-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2011 UkSim 13th International Conference on Computer Modelling and Simulation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/UKSIM.2011.88","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In this paper, we study general K-queue first-in-first-out homogeneous fork/join queueing (HFJ) systems for any K ≥ 2. We simulate and analyze an upper-bound for the mean response time that we denote by T[K]. The upper-bound uses a relatively tiny-scale system to predict the performance of a huge-scale system. It is evaluated for 10-million queues on a regular HP-PC with Intel i7-860 for three different HFJ cases. The maximum time is 16 minutes, which is only about 0.01% of the full system simulation time. We show that it is fast, close, economical and consistent by comparison and analysis.