Overcoming the power wall by exploiting inexactness and emerging COTS architectural features: Trading precision for improving application quality

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.

[1]  Krishna V. Palem,et al.  Compilers, architectures and synthesis for embedded computing: retrospect and prospect , 2010, CASES '10.

[2]  Wolfgang E. Nagel,et al.  Power measurement techniques on standard compute nodes: A quantitative comparison , 2013, 2013 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).

[3]  Krishna V. Palem,et al.  Ultra Low-energy Computing via Probabilistic Algorithms and Devices: CMOS Device Primitives and the Energy-Probability Relationship , 2004 .

[4]  Lingamneni Avinash,et al.  Ten Years of Building Broken Chips: The Physics and Engineering of Inexact Computing , 2013, TECS.

[5]  John Augustine,et al.  Opportunities for energy efficient computing: A study of inexact general purpose processors for high-performance and big-data applications , 2015, 2015 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[6]  Jack J. Dongarra,et al.  Using Mixed Precision for Sparse Matrix Computations to Enhance the Performance while Achieving 64-bit Accuracy , 2008, TOMS.

[7]  Krishna V. Palem,et al.  Computational Proof as Experiment: Probabilistic Algorithms from a Thermodynamic Perspective , 2003, Verification: Theory and Practice.

[8]  David M. Brooks,et al.  Energy characterization and instruction-level energy model of Intel's Xeon Phi processor , 2013, International Symposium on Low Power Electronics and Design (ISLPED).

[9]  John Shalf,et al.  Exascale Computing Trends: Adjusting to the "New Normal"' for Computer Architecture , 2013, Computing in Science & Engineering.

[10]  John P Holdren,et al.  National Strategic Computing Initiative Strategic Plan , 2016 .

[11]  Krishna V. Palem,et al.  Energy aware computing through probabilistic switching: a study of limits , 2005, IEEE Transactions on Computers.

[12]  Sally A. McKee,et al.  Hitting the memory wall: implications of the obvious , 1995, CARN.

[13]  John E. Dennis,et al.  Numerical methods for unconstrained optimization and nonlinear equations , 1983, Prentice Hall series in computational mathematics.

[14]  Jan Reineke,et al.  Ascertaining Uncertainty for Efficient Exact Cache Analysis , 2017, CAV.

[15]  Tim N. Palmer,et al.  Modelling: Build imprecise supercomputers , 2015, Nature.

[16]  James Demmel,et al.  Precimonious: Tuning assistant for floating-point precision , 2013, 2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC).