{"title":"A Configuration Locking Technique to Reduce the Configuration Overhead of Run-Time Reconfigurable Devices","authors":"Yang Qu, J. Soininen, J. Nurmi","doi":"10.1109/ISSOC.2007.4427423","DOIUrl":null,"url":null,"abstract":"Run-time reconfigurable logic is an interesting design alterative in SoC design because it can simultaneously provide high performance and flexibility. However, its configuration overhead can largely decrease the system performance. In this work, we present a novel configuration locking technique to reduce the overhead. The idea is to lock at run-time a number of frequently used tasks on the configuration memory so that they cannot be evicted by other tasks. A number of real applications were used to validate the approach. The results show that using proper amount of resources to lock frequently used tasks can significantly improve the performance.","PeriodicalId":244119,"journal":{"name":"2007 International Symposium on System-on-Chip","volume":"56 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2007-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2007 International Symposium on System-on-Chip","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISSOC.2007.4427423","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
Run-time reconfigurable logic is an interesting design alterative in SoC design because it can simultaneously provide high performance and flexibility. However, its configuration overhead can largely decrease the system performance. In this work, we present a novel configuration locking technique to reduce the overhead. The idea is to lock at run-time a number of frequently used tasks on the configuration memory so that they cannot be evicted by other tasks. A number of real applications were used to validate the approach. The results show that using proper amount of resources to lock frequently used tasks can significantly improve the performance.