S. Shimizu, T. Kihara, Y. Arakawa, N. Yamanaka, K. Shiba
{"title":"A prototype of a dynamically reconfigurable processor based off-loading engine for accelerating the shortest path calculation with GNU zebra","authors":"S. Shimizu, T. Kihara, Y. Arakawa, N. Yamanaka, K. Shiba","doi":"10.1109/HSPR.2008.4734433","DOIUrl":null,"url":null,"abstract":"A hardware off-loading engine to speed up the shortest path calculation in OSPF (open shortest path first) has been developed. The developed system is co-designed with both hardware and software to optimize an architecture of a router for highly functional traffic engineering (TE). To speed up the shortest path calculation, we employ a dynamically reconfigurable processor, IPFlex DAPDNA-2, as a hardware off-loader, and newly structured a novel high-speed parallel shortest path algorithm, called MPSA (Multi-route Parallel Search Algorithm). The proposed algorithm consists of simple processing, in which multiple paths are simultaneously searched by multiple processor element (PE) of DAPDNA-2. Therefore, it reduces the execution time of shortest path calculation to 2.8% compared with the popular shortest path algorithm, Dijkstrapsilas algorithm. Our prototype works together with a famous software-based router, GNU Zebra, on commodity Linux PC. The proposed architecture and prototype system can be applied to future network sophisticated TE.","PeriodicalId":130484,"journal":{"name":"2008 International Conference on High Performance Switching and Routing","volume":"48 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 International Conference on High Performance Switching and Routing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HSPR.2008.4734433","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
A hardware off-loading engine to speed up the shortest path calculation in OSPF (open shortest path first) has been developed. The developed system is co-designed with both hardware and software to optimize an architecture of a router for highly functional traffic engineering (TE). To speed up the shortest path calculation, we employ a dynamically reconfigurable processor, IPFlex DAPDNA-2, as a hardware off-loader, and newly structured a novel high-speed parallel shortest path algorithm, called MPSA (Multi-route Parallel Search Algorithm). The proposed algorithm consists of simple processing, in which multiple paths are simultaneously searched by multiple processor element (PE) of DAPDNA-2. Therefore, it reduces the execution time of shortest path calculation to 2.8% compared with the popular shortest path algorithm, Dijkstrapsilas algorithm. Our prototype works together with a famous software-based router, GNU Zebra, on commodity Linux PC. The proposed architecture and prototype system can be applied to future network sophisticated TE.