{"title":"考虑概率和任务迁移的软实时系统节能调度","authors":"Ying Li, J. Niu, Xiang Long, Meikang Qiu","doi":"10.1109/ComComAp.2014.7017212","DOIUrl":null,"url":null,"abstract":"The main challenges for embedded real-time systems which use battery as their power supply are both to satisfy the requirements of real-time systems and minimize the energy consumption. This paper studies the energy saving problem for DAG (Directed Acyclic Graph) tasks in soft real-time systems with heterogeneous multicore processors. Since soft real-time systems can tolerate occasional time violations and tasks are completed before deadlines with a given probability, this paper proposes a novel processor and voltage assignment scheme - Adaptive Processor and Voltage Assignment with Probability (APVAP) to realize the minimum energy consumption which can satisfy the requirements of time constraints under the given probability. Most of previous work focuses on multicore processor task assignment for predecessor and successor (P-S) tasks. However, this paper introduced affinity to indicate successor tasks can be re-allocated to more appropriate cores according to task features and workload. Besides, this paper introduces the concept of data migration energy (DME) to compute the transmission energy when a task is migrated to a different core and adopts Ratio between Time and Energy (RTE) to determine the most suitable tasks for migration to reduce energy consumption at the cost of execution time. The experimental results demonstrate that our approach outperforms state-of-the-art algorithms in this field (maximum improvement of 30.7%).","PeriodicalId":422906,"journal":{"name":"2014 IEEE Computers, Communications and IT Applications Conference","volume":"83 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"Energy efficient scheduling with probability and task migration considerations for soft real-time systems\",\"authors\":\"Ying Li, J. Niu, Xiang Long, Meikang Qiu\",\"doi\":\"10.1109/ComComAp.2014.7017212\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The main challenges for embedded real-time systems which use battery as their power supply are both to satisfy the requirements of real-time systems and minimize the energy consumption. This paper studies the energy saving problem for DAG (Directed Acyclic Graph) tasks in soft real-time systems with heterogeneous multicore processors. Since soft real-time systems can tolerate occasional time violations and tasks are completed before deadlines with a given probability, this paper proposes a novel processor and voltage assignment scheme - Adaptive Processor and Voltage Assignment with Probability (APVAP) to realize the minimum energy consumption which can satisfy the requirements of time constraints under the given probability. Most of previous work focuses on multicore processor task assignment for predecessor and successor (P-S) tasks. However, this paper introduced affinity to indicate successor tasks can be re-allocated to more appropriate cores according to task features and workload. Besides, this paper introduces the concept of data migration energy (DME) to compute the transmission energy when a task is migrated to a different core and adopts Ratio between Time and Energy (RTE) to determine the most suitable tasks for migration to reduce energy consumption at the cost of execution time. The experimental results demonstrate that our approach outperforms state-of-the-art algorithms in this field (maximum improvement of 30.7%).\",\"PeriodicalId\":422906,\"journal\":{\"name\":\"2014 IEEE Computers, Communications and IT Applications Conference\",\"volume\":\"83 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2014-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2014 IEEE Computers, Communications and IT Applications Conference\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ComComAp.2014.7017212\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 IEEE Computers, Communications and IT Applications Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ComComAp.2014.7017212","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Energy efficient scheduling with probability and task migration considerations for soft real-time systems
The main challenges for embedded real-time systems which use battery as their power supply are both to satisfy the requirements of real-time systems and minimize the energy consumption. This paper studies the energy saving problem for DAG (Directed Acyclic Graph) tasks in soft real-time systems with heterogeneous multicore processors. Since soft real-time systems can tolerate occasional time violations and tasks are completed before deadlines with a given probability, this paper proposes a novel processor and voltage assignment scheme - Adaptive Processor and Voltage Assignment with Probability (APVAP) to realize the minimum energy consumption which can satisfy the requirements of time constraints under the given probability. Most of previous work focuses on multicore processor task assignment for predecessor and successor (P-S) tasks. However, this paper introduced affinity to indicate successor tasks can be re-allocated to more appropriate cores according to task features and workload. Besides, this paper introduces the concept of data migration energy (DME) to compute the transmission energy when a task is migrated to a different core and adopts Ratio between Time and Energy (RTE) to determine the most suitable tasks for migration to reduce energy consumption at the cost of execution time. The experimental results demonstrate that our approach outperforms state-of-the-art algorithms in this field (maximum improvement of 30.7%).