TY - GEN
T1 - Overcoming the power wall by exploiting inexactness and emerging COTS architectural features
T2 - 29th IEEE International System on Chip Conference, SOCC 2016
AU - Fagan, Mike
AU - Schlachter, Jeremy
AU - Yoshii, Kazutomo
AU - Leyffer, Sven
AU - Palem, Krishna
AU - Snir, Marc
AU - Wild, Stefan M.
AU - Enz, Christian
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/7/2
Y1 - 2016/7/2
N2 - Energy and power consumption are major limitations to continued scaling of computing systems. Inexactness where the quality of the solution can be traded for energy savings has been proposed as a counterintuitive approach to overcoming those limitation. However, in the past, inexactness has been necessitated the need for highly customized or specialized hardware. In order to move away from customization, in earlier work [1], it was shown that by interpreting precision in the computation to be the parameter to trade to achieve inexactness, weather prediction and page rank could both benefit in terms of yielding energy savings through reduced precision, while preserving the quality of the application. However, this required representations of numbers that were not readily available on commercial off-the-shelf (COTS) processors. In this paper, we provide opportunities for extending the notion of trading precision for energy savings into the world COTS. We provide a model and analyze the opportunities and behavior of all three IEEE compliant precision values available on COTS processors: (i) double (ii) single, and (iii) half. Through measurements, we show through a limit study energy savings in going from double precision to half precision are a factor of 3.98.
AB - Energy and power consumption are major limitations to continued scaling of computing systems. Inexactness where the quality of the solution can be traded for energy savings has been proposed as a counterintuitive approach to overcoming those limitation. However, in the past, inexactness has been necessitated the need for highly customized or specialized hardware. In order to move away from customization, in earlier work [1], it was shown that by interpreting precision in the computation to be the parameter to trade to achieve inexactness, weather prediction and page rank could both benefit in terms of yielding energy savings through reduced precision, while preserving the quality of the application. However, this required representations of numbers that were not readily available on commercial off-the-shelf (COTS) processors. In this paper, we provide opportunities for extending the notion of trading precision for energy savings into the world COTS. We provide a model and analyze the opportunities and behavior of all three IEEE compliant precision values available on COTS processors: (i) double (ii) single, and (iii) half. Through measurements, we show through a limit study energy savings in going from double precision to half precision are a factor of 3.98.
UR - http://www.scopus.com/inward/record.url?scp=85019075871&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85019075871&partnerID=8YFLogxK
U2 - 10.1109/SOCC.2016.7905477
DO - 10.1109/SOCC.2016.7905477
M3 - Conference contribution
AN - SCOPUS:85019075871
T3 - International System on Chip Conference
SP - 241
EP - 246
BT - Proceedings - 29th IEEE International System on Chip Conference, SOCC 2016
A2 - Bhatia, Karan
A2 - Alioto, Massimo
A2 - Zhao, Danella
A2 - Marshall, Andrew
A2 - Sridhar, Ramalingam
PB - IEEE Computer Society
Y2 - 6 September 2016 through 9 September 2016
ER -