Value and Energy Optimizing Dynamic Resource Allocation in Many-Core HPC Systems
Singh, Amit Kumar, DZIURZANSKI, Piotr and Indrusiak, Leandro Soares (2015) Value and Energy Optimizing Dynamic Resource Allocation in Many-Core HPC Systems. In: Cloud Computing Technology and Science (CloudCom), 2015 IEEE 7th International Conference on, 30 Nov.-3 Dec. 2015, Vancouver, Canada.
Full text not available from this repository. (Request a copy)Abstract or description
The conventional approaches to reduce the energy consumption of high performance computing (HPC) data centers focus on consolidation and dynamic voltage and frequency scaling (DVFS). Most of these approaches consider independent tasks (or jobs) and do not jointly optimize for energy and value. In this paper, we propose DVFS-aware profiling and non-profiling based approaches that use design-time profiling results and perform all the computations at run-time, respectively. The profiling based approach is suitable for the scenarios when the jobs or their structure is known at design-time, otherwise, the non-profiling based approach is more suitable. Both the approaches consider jobs containing dependent tasks and exploit efficient allocation combined with identification of voltage/frequency levels of used system cores to jointly optimize value and energy. Experiments show that the proposed approaches reduce energy consumption by 15% when compared to existing approaches while achieving significant amount of value and reducing percentage of rejected jobs leading to zero value.
Item Type: | Conference or Workshop Item (Paper) |
---|---|
Faculty: | Previous Faculty of Computing, Engineering and Sciences > Computing |
Event Title: | Cloud Computing Technology and Science (CloudCom), 2015 IEEE 7th International Conference on |
Event Location: | Vancouver, Canada |
Event Dates: | 30 Nov.-3 Dec. 2015 |
Depositing User: | Piotr DZIURZANSKI |
Date Deposited: | 31 Oct 2016 10:16 |
Last Modified: | 24 Feb 2023 13:44 |
URI: | https://eprints.staffs.ac.uk/id/eprint/2655 |
Actions (login required)
View Item |