Abstract
MuMMI (Multiple Metrics Modeling Infrastructure) is an infrastructure that facilitates systematic measurement, modeling, and prediction of performance and power consumption, and performance-power tradeoffs and optimization for parallel systems. MuMMI builds upon three existing frameworks: Prophesy for performance modeling and prediction of parallel applications, PAPI for hardware performance counter monitoring, and PowerPack for power measurement and profiling. In this paper, we present the MuMMI framework, which consists of an Instrumentor, Databases and Analyzer. The MuMMI Instrumentor provides automatic performance and power data collection and storage with low overhead. The MuMMI Databases store performance, power and energy consumption and hardware performance counters’ data with different CPU frequency settings for modeling and comparison. The MuMMI Analyzer entails performance and power modeling and performance-power tradeoff and optimizations. For case studies, we apply MuMMI to a parallel earthquake simulation to illustrate building performance and power models of the application and optimizing its performance and power for energy efficiency.
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
References
Browne, S., Dongarra, J., Garner, N., Ho, G., Mucci, P.: A portable programming interface for performance evaluation on modern processors. Int. J. High-Perform. Comput. Appl. 14(2), 189–204 (2000)
Curtis-Maury, M., Dzierwa, J., et al.: Online power-performance adaptation of multithreaded programs using hardware event-based prediction. In: The International Conference on Supercomputing, Santa Fe (2006)
Ge, R., Feng, X., Cameron, K.W.: Performance-constrained distributed DVS scheduling for scientific applications on power-aware clusters. In: IEEE/ACM SC 2005, Seattle (2005)
Ge, R., Feng, X., Song, S., et al.: PowerPack: energy profiling and analysis of high-performance systems and applications. IEEE Trans. Parallel Distrib. Syst. 21(5), 658–671 (2010)
Integrated Performance Monitoring (IPM). http://ipm-hpc.sourceforge.net/
Kappiah, N., Freeh, V., Lowenthal, D.: Just in time dynamic voltage scaling: exploiting inter-node slack to save energy in MPI programs. In: The 2005 ACM/IEEE Conference on Supercomputing (SC05), Seattle (2005)
Li, D., de Supinski, B., Schulz, M., Cameron, K., Nikolopoulos, D.: Hybrid MPI/OpenMP power-aware computing. In: Proceedings of the 24th IEEE International Conference on Parallel & Distributed Processing Symposium, Atlanta, May 2010
Lively, C., Wu, X., Taylor, V., Moore, S., Chang, H., Cameron, K.: Energy and performance characteristics of different parallel implementations of scientific applications on multicore systems. Int. J. High Perform. Comput. Appl. (IJHPCA) 25(3), 342–350 (2011)
Lively, C., Wu, X., Taylor, V., Moore, S., Chang, H., Su, C., Cameron, K.: Power-aware predictive models of hybrid (MPI/OpenMP) scientific applications on multicore systems. Comput. Sci. Res. Dev. 27(4), 245–253 (2012). Springer
Lively, C., Taylor, V., Wu, X., Chang, H., Su, C., Cameron, K., Moore, S., Terpstra, D.: E-AMOM: an energy-aware modeling and optimization methodology for scientific applications on multicore systems. In: International Conference on Energy-Aware High Performance Computing, Dresden, Sep 2–3, 2013
Multiple Metrics Modeling Infrastructure (MuMMI) project. http://www.mummi.org
NVIDIA, NVIDIA’s Management Library (NVML) API Reference Manual (2012)
PAPI (Performance API). http://icl.cs.utk.edu/papi/
Rotem, E., Naveh, A., Rajwan, D., Ananthakrishnan, A., Weissmann, E.: Power-management architecture of the intel microarchitecture code-named sandy bridge. IEEE Micro. 32(2), 20–27 (2012)
Score-P, Scalable Performance Measurement Infrastructure for Parallel Codes. http://www.vi-hps.org/projects/score-p/
Singh, K., Bhadhauria, M., McKee, S.A.: Real time power estimation and thread scheduling via performance counters. In: Proceedings of the Workshop on Design, Architecture, and Simulation of Chip Multi-Processors, Lake Como, Nov 2008
SOAP (Simple Object Access Protocol). http://www.w3.org/TR/soap/
SystemG at Virginia Tech. http://www.cs.vt.edu/facilities/systemg
TAU (Tuning and Analysis Utilities). http://www.cs.uoregon.edu/research/tau
Taylor, V., Wu, X., Geisler, J., Stevens, R.: Using kernel couplings to predict parallel application performance. In: Proceedings of the 11th IEEE International Symposium on High-Performance Distributed Computing (HPDC 2002), Edinburgh, 24–26 July 2002
Taylor, V., Wu, X., Stevens, R.: Prophesy: an infrastructure for performance analysis and modeling of parallel and grid applications. ACM SIGMETRICS Perform. Eval. Rev. 30(4), 13–18 (2003)
The SCEC/USGS Spontanous Rupture Code Verification Project. http://scecdata.usc.edu/cvws
Wu, X., Taylor, V., Stevens, R.: Design and implementation of prophesy automatic instrumentation and data entry system. In: The 13th International Conference on Parallel and Distributed Computing and Systems (PDCS2001), Anaheim (2001)
Wu, X., Taylor, V., et al.: Design and development of prophesy performance database for distributed scientific applications. In: Proceedings of the 10th SIAM Conference on Parallel Processing for Scientific Computing, Virginia (2001)
Wu, X., Taylor, V., Geisler, J., Stevens, R.: Isocoupling: reusing coupling values to predict parallel application performance. In: 18th International Parallel and Distributed Processing Symposium (IPDPS2004), Santa Fe (2004)
Wu, X., Duan, B., Taylor, V.: Parallel simulations of dynamic earthquake rupture along geometrically complex faults on CMP systems. J. Algorithm Comput. Technol. 5(2), 313–340 (2011)
Wu, X., Duan, B., Taylor, V.: Parallel earthquake simulations on large-scale multicore supercomputers, (Book Chapter). In: Furht, B., Escalante, A. (eds.) Handbook of Data Intensive Computing. Springer, New York (2011)
Yoshii, K., Iskra, K., Gupta, R., Beckman, P., Vishwanath, V., Yu, C., Coghlan, S.: Evaluating power monitoring capabilities on IBM Blue Gene/P and Blue Gene/Q. In: IEEE Conference on Cluster Computing, Beijing (2012)
Acknowledgements
This work is supported by NSF grants CNS-0911023, CNS-0910899, and CNS-0910784.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Wu, X. et al. (2014). MuMMI: Multiple Metrics Modeling Infrastructure. In: Knüpfer, A., Gracia, J., Nagel, W., Resch, M. (eds) Tools for High Performance Computing 2013. Springer, Cham. https://doi.org/10.1007/978-3-319-08144-1_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-08144-1_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-08143-4
Online ISBN: 978-3-319-08144-1
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)