A survey of system level power management schemes in the dark-silicon era for many-core architectures

Ofori-Attah, E., Wang, X. and Opoku Agyeman, M. (2018) A survey of system level power management schemes in the dark-silicon era for many-core architectures. EAI Endorsed Transactions on Industrial Networks and Intelligent Systems. 5(15) 2410-0218.

Information

Item Type:	Article
Abstract:	Power consumption in Complementary Metal Oxide Semiconductor (CMOS) technology has escalated to a point that only a fractional part of many-core chips can be powered-on at a time. Fortunately, this fraction can be increased at the expense of performance through the dark-silicon solution. However, with many-core integration set to be heading towards its thousands, power consumption and temperature increases per time, meaning the number of active nodes must be reduced drastically. Therefore, optimized techniques are demanded for continuous advancement in technology. Existing eﬀorts try to overcome this challenge by activating nodes from diﬀerent parts of the chip at the expense of communication latency. Other eﬀorts on the other hand employ run-time power management techniques to manage the power performance of the cores trading-oﬀ performance for power. We found out that, for a signiﬁcant amount of power to saved and high temperature to be avoided, focus should be on reducing the power consumption of all the on-chip components. Especially, the memory hierarchy and the interconnect. Power consumption can be minimized by, reducing the size of high leakage power dissipating elements, turning-oﬀ idle resources and integrating power saving materials.
Uncontrolled Keywords:	Dark-silicon, many-core, multi-core, NoC, low power NoC architectures, power budgeting
Creators:	Ofori-Attah, Emmanuel, Wang, Xiaohang and Opoku Agyeman, Michael
Faculties, Divisions and Institutes:	Faculties > Faculty of Arts, Science & Technology > Computing
Date:	19 September 2018
Date Type:	Publication
Journal or Publication Title:	EAI Endorsed Transactions on Industrial Networks and Intelligent Systems
Volume:	5
Number:	15
Language:	English
DOI:	https://doi.org/10.4108/eai.19-9-2018.155569
ISSN:	2410-0218
Status:	Published / Disseminated
Refereed:	Yes
URI:	http://nectar.northampton.ac.uk/id/eprint/10617

Actions (login required)

Edit Item

Altmetric