A Performance Tool for Earth System Models Development

Abstract:

We present a performance tool for Earth system models development to aid in analyzing the performance of the climate modeling applications. It is difficult for existing tools to handle with the complex, coupled structure and the long execution time of models. Our performance tool implements rapid analysis based on statistical sampling and grouping aggregation the calling relationship and the actual computing resource consumption excluding waiting losses. Using this tool, we study an ocean model POP in short-term sampling and analyze its scaling bottleneck and acceleration trend. The measuring results of its entire execution prove our predictions on the scaling efficiencies.

• Periodical: Advanced Materials Research (Volumes 756-759)
• Pages: 3814-3820
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.756-759.3814
• Online since: September 2013
• Authors: Hang Li, Zhong Zhi Luan
• Keywords: Earth System Models, Parallel Computing, Performance Analysis

References

[1] C. Hill, C. DeLuca, V. Balaji, M. Suarez, and A. Da Silva, The architecture of the Earth System Modeling Framework, Computing in Science and Engineering, 6(1), p.18–28, (2004).

DOI: 10.1109/mcise.2004.1255817

[2] D. C. Arnold, D. H. Ahn, B. R. de Supinski, G. L. Lee, B. P. Miller, and M. Schulz, Stack trace analysis for large scale debugging, Parallel and Distributed Processing Symposium 2007(IPDPS 2007), IEEE International, Madison, p.1–10, (2007).

DOI: 10.1109/ipdps.2007.370254

[3] N. R. Tallent, J. M. Mellor-Crummey, M. Franco, R. Landrum, and L. Adhianto, Scalable fine-grained call path tracing, " Proceedings of the international conference on Supercomputing, ICS , 11, ACM, New York, USA, pp.63-74, (2011).

DOI: 10.1145/1995896.1995908

[4] W. E. Nagel, A. Arnold, M. Weber, H. C. Hoppe, and K. Solchenbach, VAMPIR: Visualization and analysis of MPI resources, Supercomputer, 12(1): 69-80, (1996).

[5] S. S. Shende and A. D. Malony, The TAU parallel performance system, International Journal of High Performance Computing Applications, 20(2): 287–311, (2006).

DOI: 10.1177/1094342006064482

[6] A. D. Malony, S. Shende, A. Morris, S. Biersdor, W. Spear, K. Huck, and A. Nataraj, Evolution of a parallel performance system, Tools for High Performance Computing, Springer, Berlin Heidelberg, p.169–190, (2008).

DOI: 10.1007/978-3-540-68564-7_11

[7] B. P. Miller, M. D. Callaghan, J. M. Cargille, J. K. Hollingsworth, R. B. Irvin, K. L. Karavanic, K. Kunchithapadam, and T. Newhall, The Paradyn parallel performance measurement tool, Computer, IEEE Computer Society, 28(11): 37–46, (1995).

DOI: 10.1109/2.471178

[8] N. R. Tallent and J. Mellor-Crummey, Effective performance measurement and analysis of multithreaded applications, Technical report, Rice University, August (2008).

[9] N. R. Tallent, J. M. Mellor-Crummey, L. Adhianto, M. W. Fagan, and M. Krentel, Diagnosing performance bottlenecks in emerging petascale applications, In Proceedings of the Conference on High Performance Computing Networking, ACM, New York, USA, p.1–11, (2009).

DOI: 10.1145/1654059.1654111

[10] Earth System Modeling Framework, http: /www. earthsystemmodeling. org/, (2011).

[11] P. H. Worley, A. P. Craig, J. M. Dennis, A. A. Mirin, M. A. Taylor, and M. Vertenstein, Performance of the Community Earth System Model, " In Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC , 11), ACM, Seatle, WA, USA, pp.1-11, (2011).

DOI: 10.1145/2063384.2063457

[12] M. Blackmon, B. Boville, F. Bryan, P. Gent, J. Kiehl, G. Bonan, et al., The Community Climate System Model, Bulletin of the American Meteorological Society, American Meteorological Society, Boston, MA, 82(11): 2357–2376, (2001).

DOI: 10.1175/1520-0477(2001)082<2357:tccsm>2.3.co;2

[13] V. Balaji, FMS: The GFDL Flexible Modelling System. http: /www. gfdl. noaa. gov/fms, (2004).

[14] N. R. Tallent, L. Adhianto, and J. M. Mellor-Crummey, Scalable identification of load imbalance in parallel executions using call path profiles, " In Proceedings of 2010 International Conference for High Performance Computing, Networking, Storage and Analysis (SC , 10), ACM, New Orleans, LA, pp.1-11, (2010).

DOI: 10.1109/sc.2010.47

[15] R. L. Henderson, Job scheduling under the portable batch system,. In Job Scheduling Strategies for Parallel Processing, Springer-Verlag, p.279–294, (1995).

DOI: 10.1007/3-540-60153-8_34

[16] A.B. Yoo, M. A. Jette, and M. Grondona, SLURM: Simple Linux Utility for Resource Management, In Job Scheduling Strategies for Parallel Processing, SpringerVerlag, pp.44-60, (2003).

DOI: 10.1007/10968987_3

[17] Intel Corporation. Intel VTune performance analyzer. http: /www. intel. com/software/products/vtune.

DOI: 10.1109/icisc.2018.8399107

[18] S. L. Graham, P. B. Kessler, and M. K. McKusick, Gprof: A call graph execution profiler, " In Proceedings of the 1982 SIGPLAN Symposium on Compiler Construction (SIGPLAN , 82), ACM Press, New York, NY, USA, p.120–126, (1982).

DOI: 10.1145/800230.806987