{"title":"A VLIW low power Java processor for embedded applications","authors":"A. C. S. Beck, L. Carro","doi":"10.1145/1016568.1016614","DOIUrl":null,"url":null,"abstract":"This paper presents a pioneer VLIW architecture of a native Java processor. We show that, thanks to the specific stack architecture and to the use of the VLIW technique, one is able to obtain a meaningful reduction of power dissipation, with small area overhead, when compared to other ways of executing Java in hardware. The underlying technique is based on the reuse of memory access instructions, hence reducing power during memory or cache accesses. The architecture is validated for some complex embedded applications like IMDCT computation and other data processing benchmarks.","PeriodicalId":275811,"journal":{"name":"Proceedings. SBCCI 2004. 17th Symposium on Integrated Circuits and Systems Design (IEEE Cat. No.04TH8784)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2004-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"30","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. SBCCI 2004. 17th Symposium on Integrated Circuits and Systems Design (IEEE Cat. No.04TH8784)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/1016568.1016614","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 30
Abstract
This paper presents a pioneer VLIW architecture of a native Java processor. We show that, thanks to the specific stack architecture and to the use of the VLIW technique, one is able to obtain a meaningful reduction of power dissipation, with small area overhead, when compared to other ways of executing Java in hardware. The underlying technique is based on the reuse of memory access instructions, hence reducing power during memory or cache accesses. The architecture is validated for some complex embedded applications like IMDCT computation and other data processing benchmarks.