Leonardo Kunrath, Carlos Becker Westphall, F. Koch
{"title":"Towards Advance Reservation in Large-Scale Grids","authors":"Leonardo Kunrath, Carlos Becker Westphall, F. Koch","doi":"10.1109/ICONS.2008.41","DOIUrl":null,"url":null,"abstract":"Advance resource reservations allow clients to have exclusive access to Grid resources. The goal is to deliver higher levels of Quality of Service. However, in practice, the support for advance reservations in Grids has been restricted to processor reservations on clusters or bandwidth reservation on high speed LANs. This is a shortcoming in current Grid architectures. In this paper, we present an analysis of current resource reservation strategies and propose a new approach to extend advance reservations to almost any Grid node. We validate our approach through the implementation of a resource scheduler and two highly scalable scheduling algorithms. We also verify the algorithms performance through simulation. The results show that the resource utilization rate can be over 90% depending on the number of reservations.","PeriodicalId":127732,"journal":{"name":"Third International Conference on Systems (icons 2008)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Third International Conference on Systems (icons 2008)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICONS.2008.41","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
Advance resource reservations allow clients to have exclusive access to Grid resources. The goal is to deliver higher levels of Quality of Service. However, in practice, the support for advance reservations in Grids has been restricted to processor reservations on clusters or bandwidth reservation on high speed LANs. This is a shortcoming in current Grid architectures. In this paper, we present an analysis of current resource reservation strategies and propose a new approach to extend advance reservations to almost any Grid node. We validate our approach through the implementation of a resource scheduler and two highly scalable scheduling algorithms. We also verify the algorithms performance through simulation. The results show that the resource utilization rate can be over 90% depending on the number of reservations.