{"title":"一种基于动态电压缩放的MPSoC调度新技术","authors":"Chang-Woo Park, Kyung-Woo Noh, Seok-Yoon Kim","doi":"10.1109/SOCDC.2008.4815687","DOIUrl":null,"url":null,"abstract":"The DVS(Dynamic Voltage Scaling) technique is the method to reduce the dynamic energy consumption. Using slack times, it extends the execution time of big load operations by changing frequency and voltage of variable voltage processors. Many researches have been going on to control the energy consumption of the processors and the data transmission among processors by controlling the bandwidth to reduce the energy consumption of the entire system. Since operations in multiprocessor systems have the data dependency between processors, however, the DVS techniques devised for single processors are not suitable to improve the energy efficiency of multiprocessor systems. We propose a new scheduling algorithm based on DVS for increasing energy efficiency of multiprocessor systems. The proposed DVS-based scheduling algorithm can improve the energy efficiency of entire systems because it controls frequency and voltages while considering the data dependency among processors.","PeriodicalId":405078,"journal":{"name":"2008 International SoC Design Conference","volume":"44 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"A new scheduling technique based on Dynamic Voltage Scaling for MPSoC\",\"authors\":\"Chang-Woo Park, Kyung-Woo Noh, Seok-Yoon Kim\",\"doi\":\"10.1109/SOCDC.2008.4815687\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The DVS(Dynamic Voltage Scaling) technique is the method to reduce the dynamic energy consumption. Using slack times, it extends the execution time of big load operations by changing frequency and voltage of variable voltage processors. Many researches have been going on to control the energy consumption of the processors and the data transmission among processors by controlling the bandwidth to reduce the energy consumption of the entire system. Since operations in multiprocessor systems have the data dependency between processors, however, the DVS techniques devised for single processors are not suitable to improve the energy efficiency of multiprocessor systems. We propose a new scheduling algorithm based on DVS for increasing energy efficiency of multiprocessor systems. The proposed DVS-based scheduling algorithm can improve the energy efficiency of entire systems because it controls frequency and voltages while considering the data dependency among processors.\",\"PeriodicalId\":405078,\"journal\":{\"name\":\"2008 International SoC Design Conference\",\"volume\":\"44 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2008-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2008 International SoC Design Conference\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/SOCDC.2008.4815687\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 International SoC Design Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SOCDC.2008.4815687","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A new scheduling technique based on Dynamic Voltage Scaling for MPSoC
The DVS(Dynamic Voltage Scaling) technique is the method to reduce the dynamic energy consumption. Using slack times, it extends the execution time of big load operations by changing frequency and voltage of variable voltage processors. Many researches have been going on to control the energy consumption of the processors and the data transmission among processors by controlling the bandwidth to reduce the energy consumption of the entire system. Since operations in multiprocessor systems have the data dependency between processors, however, the DVS techniques devised for single processors are not suitable to improve the energy efficiency of multiprocessor systems. We propose a new scheduling algorithm based on DVS for increasing energy efficiency of multiprocessor systems. The proposed DVS-based scheduling algorithm can improve the energy efficiency of entire systems because it controls frequency and voltages while considering the data dependency among processors.