Abstract
We describe and validate a complete hardware/software environment for power consumption analysis of PCIe-based accelerators, using the Intel Xeon Phi co-processor as the target platform. Our environment is flexible and affordable—based on commodity instrumentation—and provides both accuracy and transparency for the user, which enables an easy instrumentation of existing codes from the power consumption perspective. We present empirical power traces for two well known scientific codes (LINPACK and libflame) that give insights not only on the benefits of the presented environment, but also on the power profile of the Intel Xeon Phi co-processor under different workloads.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
(1999) Standard commands for programmable instruments, vol 1. SCPI Consortium
(2013) How a single chip causes massive power bills. GPUSimPow: a GPGPU power simulator. In: ISPASS 2013, IEEE, pp 97–106
(2013) Intel Xeon Phi Coprocessor Datasheet. Intel Corporation
Barrachina S, Barreda M, Catalán S, Dolz MF, Fabregat G, Mayo R, Quintana-Ortí ES (2013) An integrated framework for power-performance analysis of parallel scientific workloads. In: 3rd int conf on smart grids, green communications and IT energy-aware technologies, pp 114–119
Chan E, Quintana-Ortí ES, Quintana-Ortí G, van de Geijn R (2007) SuperMatrix out-of-order scheduling of matrix operations for SMP and multi-core architectures. SPAA ’07. San Diego, CA, USA, pp 116–125
Dongarra JJ, Luszczek P, Petitet A (2003) The linpack benchmark: Past, present, and future. CC: P&E 15:2003
Jackson J (2013) Comparison and analysis of GPU energy efficiency for CUDA and OpenCL
Shao Y, Brooks D (2013) Energy characterization and instruction-level energy model of Intel’s Xeon Phi processor. In: Low power electronics and design, pp 1–6
Ukidave Y, Ziabari AK, Mistry P, Schirner G, Kaeli D (2013) Quantifying the energy efficiency of FFT on heterogeneous platforms. In: ISPASS 2013, IEEE, pp 235–244
Zee FGV (2009) libflame: the complete reference. http://www.lulu.com
Acknowledgments
This work is supported by Projects CICYT-TIN 2008/508 and 2012-32180. We thank the technical staff at DACYA for their support in setting up the measurement environment.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Igual, F.D., Jara, L.M., Gómez-Pérez, J.I. et al. A power measurement environment for PCIe accelerators. Comput Sci Res Dev 30, 115–124 (2015). https://doi.org/10.1007/s00450-014-0266-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00450-014-0266-8