pmcEDF: An Energy Efficient Procrastination Scheduler for Multi-core Mixed Criticality Systems
2022 (English)In: IEEE Int. Conf. High Perform. Comput. Commun., Int. Conf. Data Sci. Syst., Int. Conf. Smart City Int. Conf. Dependability Sensor, Cloud Big Data Syst. Applications, HPCC-DSS-SmartCity-DependSys, Institute of Electrical and Electronics Engineers Inc. , 2022, p. 727-732Conference paper, Published paper (Refereed)
Abstract [en]
The ever increasing computational demands with size, weight, energy, power and time guidelines necessitates innovative energy efficient techniques in multi-core mixed criticality applications. Energy efficient design considerations include energy optimization techniques such as dynamic-voltage-and-frequency-scaling (DVFS) and dynamic-power-management (DPM). DPM techniques of procrastination and power shutdown reduce static and dynamic energy consumption efficiently. This work presents pmcEDF, a procrastination technique that extends the Earliest Deadline First with Virtual Deadlines (EDF-VD) scheduler to support both non-shutdown and shutdown cores. pmcEDF aims to reduce static and dynamic energy consumption by ensuring high criticality job deadlines without hampering quality of service by accommodating feasible low criticality jobs. Experimental evaluation shows that the proposed algorithm provides a total energy saving of 3.61%, 4.76% and 39.16% over existing EDF/EDF-VD based techniques.
Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers Inc. , 2022. p. 727-732
Keywords [en]
energy efficiency, mixed criticality systems, multi-core, procrastination, safety criticality
National Category
Computer Engineering
Identifiers
URN: urn:nbn:se:mdh:diva-59455DOI: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00119Scopus ID: 2-s2.0-85132423231ISBN: 9781665494571 (print)OAI: oai:DiVA.org:mdh-59455DiVA, id: diva2:1678647
Conference
2021 IEEE 23rd International Conference on High Performance Computing and Communications, 7th International Conference on Data Science and Systems, 19th International Conference on Smart City and 7th International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021
2022-06-292022-06-292022-06-29Bibliographically approved