Tools for Power-Energy Modelling and Analysis of Parallel Scientific Applications

Understanding power usage in parallel workloads is crucial to develop the energy-aware software that will run in future Exascale systems. In this paper, we contribute towards this goal by introducing an integrated framework to profile, monitor, model and analyze power dissipation in parallel MPI and multi-threaded scientific applications. The framework includes an own-designed device to measure internal DC power consumption and a package offering a simple interface to interact with this design as well as commercial power meters. Combined with the instrumentation package Extrae and the graphical analysis tool Paraver, the result is a useful environment to identify sources of power inefficiency directly in the source application code. For task-parallel codes, we also offer a statistical software module that inspects the execution trace of the application to calculate the parameters of an accurate model for the global energy consumption, which can be then decomposed into the average power usage per task or the nodal power dissipated per core.

[1]  Robert A. van de Geijn,et al.  Updating an LU Factorization with Pivoting , 2008, TOMS.

[2]  Ed Anderson,et al.  LAPACK Users' Guide , 1995 .

[3]  Babak Falsafi,et al.  The HiPEAC Vision , 2010 .

[4]  Rong Ge,et al.  Green Supercomputing Comes of Age , 2008, IT Professional.

[5]  FengWu-chun,et al.  The Green500 List , 2007 .

[6]  Jesús Labarta,et al.  Symmetric Rank-k Update on Clusters of Multicore Processors with SMPSs , 2011, PARCO.

[7]  Thomas Ludwig.  Editorial for the First International Conference on Energy-Aware High Performance Computing , 2010, Computer Science - Research and Development.

[8]  Karthikeyan Sankaralingam,et al.  Dark Silicon and the End of Multicore Scaling , 2012, IEEE Micro.

[9]  Weisong Shi,et al.  SPAN: A software power analyzer for multicore computer systems , 2011, Sustain. Comput. Informatics Syst..

[10]  John Shalf,et al.  The International Exascale Software Project roadmap , 2011, Int. J. High Perform. Comput. Appl..

[11]  Dong Li,et al.  PowerPack: Energy Profiling and Analysis of High-Performance Systems and Applications , 2010, IEEE Transactions on Parallel and Distributed Systems.

[12]  Jesús Labarta,et al.  Parallelizing dense and banded linear algebra libraries using SMPSs , 2009, Concurr. Comput. Pract. Exp..

[13]  Ralf Gruber,et al.  One Joule per GFlop for BLAS2 Now , 2010 .

[14]  Thomas Ludwig,et al.  Tool Environments to Measure Power Consumption and Computational Performance , 2012 .

[15]  J. Kunkel.  HDTrace – A Tracing and Simulation Environment of Application and System Interaction , 2011 .

[16]  R. Fowler,et al.  Powermon 2: Fine-grained, Integrated Power Measurement , 2010 .

[17]  Andrew A. Chien,et al.  The future of microprocessors , 2011, Commun. ACM.

[18]  Gene H. Golub,et al.  Matrix computations , 1983 .