H. Parandeh-Afshar, A. Tootoonchian, M. Yousefpour, O. Fatemi, M. Hashemi
{"title":"A Slice-Based Automatic Hardware/Software Partitioning Heuristic","authors":"H. Parandeh-Afshar, A. Tootoonchian, M. Yousefpour, O. Fatemi, M. Hashemi","doi":"10.1109/ICM.2006.373289","DOIUrl":null,"url":null,"abstract":"In this paper, a novel level-based hardware/software partitioning heuristic has been proposed. The algorithm operates on functional blocks of designs represented as directed acyclic graphs (DAG), with the objective of minimizing the processing time under various hardware area constraints. In most existing methods, the communication overhead and the fact that the vertices mapped onto the same computing unit have less communication, is overlooked during the partitioning decision, while the proposed algorithm considers this fact during partitioning.","PeriodicalId":284717,"journal":{"name":"2006 International Conference on Microelectronics","volume":"76 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2006 International Conference on Microelectronics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICM.2006.373289","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, a novel level-based hardware/software partitioning heuristic has been proposed. The algorithm operates on functional blocks of designs represented as directed acyclic graphs (DAG), with the objective of minimizing the processing time under various hardware area constraints. In most existing methods, the communication overhead and the fact that the vertices mapped onto the same computing unit have less communication, is overlooked during the partitioning decision, while the proposed algorithm considers this fact during partitioning.