{"title":"NOC合成与ITRS预测:基于线性规划的合成的挑战","authors":"O. Hammami","doi":"10.1109/IDT.2013.6727135","DOIUrl":null,"url":null,"abstract":"Network on chip are fundamental in the performance of complex system on chip. Numerous solutions have been proposed both for regular and irregular topologies. Irregular or custom topologies present numerous benefits with regard to area/performance optimizations and can be automatically generated through NOC synthesis flows. NOC synthesis generates NOC topologies from application requirements coregraphs. NOC synthesis techniques use either exact or heuristic techniques. So far NOC synthesis techniques have not been benchmarked against ITRS roadmaps. We propose in this paper to benchmark linear programming based NOC synthesis techniques using NOCBENCH v.1.0 benchmarks. The results show that new models and techniques are needed to overcome complexity of future manycore.","PeriodicalId":446826,"journal":{"name":"2013 8th IEEE Design and Test Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"NOC synthesis vs ITRS predictions: The challenges of linear programming based synthesis\",\"authors\":\"O. Hammami\",\"doi\":\"10.1109/IDT.2013.6727135\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Network on chip are fundamental in the performance of complex system on chip. Numerous solutions have been proposed both for regular and irregular topologies. Irregular or custom topologies present numerous benefits with regard to area/performance optimizations and can be automatically generated through NOC synthesis flows. NOC synthesis generates NOC topologies from application requirements coregraphs. NOC synthesis techniques use either exact or heuristic techniques. So far NOC synthesis techniques have not been benchmarked against ITRS roadmaps. We propose in this paper to benchmark linear programming based NOC synthesis techniques using NOCBENCH v.1.0 benchmarks. The results show that new models and techniques are needed to overcome complexity of future manycore.\",\"PeriodicalId\":446826,\"journal\":{\"name\":\"2013 8th IEEE Design and Test Symposium\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2013 8th IEEE Design and Test Symposium\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/IDT.2013.6727135\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 8th IEEE Design and Test Symposium","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IDT.2013.6727135","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
NOC synthesis vs ITRS predictions: The challenges of linear programming based synthesis
Network on chip are fundamental in the performance of complex system on chip. Numerous solutions have been proposed both for regular and irregular topologies. Irregular or custom topologies present numerous benefits with regard to area/performance optimizations and can be automatically generated through NOC synthesis flows. NOC synthesis generates NOC topologies from application requirements coregraphs. NOC synthesis techniques use either exact or heuristic techniques. So far NOC synthesis techniques have not been benchmarked against ITRS roadmaps. We propose in this paper to benchmark linear programming based NOC synthesis techniques using NOCBENCH v.1.0 benchmarks. The results show that new models and techniques are needed to overcome complexity of future manycore.