Applying Genetic Algorithms to Optimize Power in Tiled SNUCA Chip Multicore Architectures

Dani , Aparna Mandke and Amrutur, Bharadwaj and Srikant, YN (2011) Applying Genetic Algorithms to Optimize Power in Tiled SNUCA Chip Multicore Architectures. In: Symposium on Applied Computing , March 21-25 2011, Taichung, Taiwan.

PDF acm_sym_app_com_2011.pdf - Published Version Restricted to Registered users only Download (432kB) | Request a copy

Abstract

We propose a novel technique for reducing the power consumed by the on-chip cache in SNUCA chip multicore platform. This is achieved by what we call a "remap table", which maps accesses to the cache banks that are as close as possible to the cores, on which the processes are scheduled. With this technique, instead of using all the available cache, we use a portion of the cache and allocate lesser cache to the application. We formulate the problem as an energy-delay (ED) minimization problem and solve it offline using a scalable genetic algorithm approach. Our experiments show up to 40% of savings in the memory sub-system power consumption and 47% savings in energy-delay product (ED).

Item Type:	Conference Proceedings
Series.:	Proceedings of the 2011 ACM Symposium on Applied Computing
Additional Information:	Copyright for this article belongs to the ACM
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	30 Jul 2012 05:02
Last Modified:	30 Jul 2012 05:02
URI:	http://eprints.iisc.ac.in/id/eprint/44872

Actions (login required)

View Item