A3 Book chapter

Dark silicon patterning: Efficient power utilization through run-time mapping




List of Authors: Anil Kanduri, Mohammad-Hashem Haghbayan, Amir M. Rahmani, Pasi Liljeberg, Axel Jantsch, Hannu Tenhunen

Publisher: Springer International Publishing

Publication year: 2017

Book title *: The Dark Side of Silicon: Energy Efficient Computing in the Dark Silicon Era

Journal name in source: The Dark Side of Silicon: Energy Efficient Computing in the Dark Silicon Era

Number of pages: 22

ISBN: 978-3-319-31594-2

eISBN: 978-3-319-31596-6

DOI: http://dx.doi.org/10.1007/978-3-319-31596-6_9


Abstract

An efficient run-time application mapping approach can considerably
enhance resource utilization and mitigate the dark silicon phenomenon.
In this chapter, we present a dark silicon aware run-time application
mapping approach that patterns active cores alongside the inactive cores
in order to evenly distribute power density across the chip. This
approach leverages dark silicon to balance the temperature of active
cores to provide higher power budget and better resource utilization,
within a safe peak operating temperature. In contrast to exhaustive
search based mapping techniques, the proposed agile heuristic approach
has a negligible run-time overhead. This patterning strategy yields a
surplus power budget of up to 17 % along with an improved throughput of
up to 21 % in comparison with other state-of-the-art run-time mapping
strategies, while the surplus budget is as high as 40 % compared to
worst case scenarios.


Last updated on 2021-24-06 at 09:39