Abstract
The Linux™ community has long been concerned with power saving measures, particularly in the mobile computing sector. Linux has been quick to leverage architectural features of microprocessors to reduce power consumption during idle cycles (and under load as we will discuss in Chap. 6) as these features have become commercially available. The High Performance Computing (HPC) community makes use of Linux on many of their platforms, but lightweight kernels (LWKs) are often used to deliver the maximum amount of performance at large scale (Red Storm and Blue Gene, for example). To achieve greater performance at scale, LWKs have a selective feature set when compared to general purpose operating systems like Linux. As a result, LWKs are a prime area for investigating opportunities for power savings, as long as performance is not affected. In the area of idle power usage, Linux serves as an established benchmark. The goal of this first experiment is to match or beat the idle current draw of Linux.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The name Catamount [1] will generally be used throughout this book unless the more specific names CVN and CNW are necessary to point out an important distinction.
- 2.
AMD Opteron 280 AMD Dual-Core Opteron 2.4 GHz 2M Cache Socket 940 OSA280FAA6CB.
- 3.
AMD Opteron Budapest 2.2 GHz socket AM2.
- 4.
CVN was enhanced to support more than two cores, the resulting Catamount version was named CNW. Unless otherwise specified all results shown after Fig. 5.2 were obtained running on CNW.
References
S.M. Kelly, R.B. Brightwell, Software architecture of the light weight kernel, Catamount, in Cray User Group, (CUG), 2005
Pittsburgh Supercomputing Center, http://www.psc.edu
J. Dongarra, J. Bunch, C. Moler, G.W. Stewart, High Performance Linpack HPL, in Technical Report CS-89-85, University of Tennessee, 1989
PALLAS, Intel, Available: http://www.intel.com/cd/software/products/asmo-na/eng/cluster/mpi/219848.htm
HPCC, DARPA, Available: http://icl.cs.utk.edu/hpcc/
DOE Energy Statistics, Department of Energy, http://www.eia.doe.gov/cneaf/electricity/epm/table5_6_a.html
C.T. Vaughan, J.P. VanDyke, S.M. Kelly, Application performance under different XT operating systems, in Cray User Group, (CUG), 2008
R. Zajcew, P. Roy, D. Black, C. Peak, P. Guedes, B. Kemp, J. LoVerso, M. Leibensperger, M. Barnett, F. Rabii, D. Netterwala, An OSF/1 UNIX for Massively Parallel Multicomputers, in Proceedings of the USENIX Technical Conference, USENIX, 1993
F. Petrini, D. Kerbyson, S. Pakin, The Case of the missing supercomputer performance: achieving optimal performance on the 8,192 Processors of ASCI Q, in Proceedings of the International Conference on High Performance Computing, Networking, Storage, and Analysis (SC), ACM/IEEE, 2003
K.B. Ferreira, R. Brightwell, P.G. Bridges, Characterizing application sensitivity to OS interference using kernel-level noise injection, in Proceedings of the International Conference on High Performance Computing, Networking, Storage, and Analysis (SC), ACM/IEEE, 2008
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2013 James H. Laros III
About this chapter
Cite this chapter
Laros III, J. et al. (2013). Reducing Power During Idle Cycles. In: Energy-Efficient High Performance Computing. SpringerBriefs in Computer Science. Springer, London. https://doi.org/10.1007/978-1-4471-4492-2_5
Download citation
DOI: https://doi.org/10.1007/978-1-4471-4492-2_5
Published:
Publisher Name: Springer, London
Print ISBN: 978-1-4471-4491-5
Online ISBN: 978-1-4471-4492-2
eBook Packages: Computer ScienceComputer Science (R0)