Visualizing More Performance Data Than What Fits on Your Screen

Conference paper

Abstract

High performance applications are composed of many processes that are executed in large-scale systems with possibly millions of computing units. A possible way to conduct a performance analysis of such applications is to register in trace files the behavior of all processes belonging to the same application. The large number of processes and the very detailed behavior that we can record about them lead to a trace size explosion both in space and time dimensions. The performance visualization of such data is very challenging because of the quantities involved and the limited screen space available to draw them all. If the amount of data is not properly treated for visualization, the analysis may give the wrong idea about the behavior registered in the traces. This paper is twofold: first, it details data aggregation techniques that are fully configurable by the user to control the level of details in both space and time dimensions; second, it presents two visualization techniques that take advantage of the aggregated data to scale. These features are part of the Viva open-source tool and framework, which is also briefly described in this paper.

Notes

Acknowledgements

This work is partially funded by the french SONGS project (ANR-11-INFRA-13) of the Agence Nationale de la Recherche (ANR). We thank Augustin Degomme for providing the sweep3D MPI traces. We also thank the organizers of the 6th International Parallel Tools Workshop for the invitation.

References

  1. 1.
    Aguilera, G., Teller, P., Taufer, M., Wolf, F.: A systematic multi-step methodology for performance analysis of communication traces of distributed applications based on hierarchical clustering. In: Parallel and Distributed Processing Symposium, 2006. IPDPS 2006. 20th International, p. 8 pp. (2006). DOI 10.1109/IPDPS.2006.1639645Google Scholar
  2. 2.
    Bolze, R., Cappello, F., Caron, E., Daydé, M., Desprez, F., Jeannot, E., Jégou, Y., Lantéri, S., Leduc, J., Melab, N., R. Namyst, G.M., Primet, P., Quetier, B., Richard, O., Talbi, E.G., Touche, I.: Grid’5000: a large scale and highly reconfigurable experimental grid testbed. International Journal of High Performance Computing Applications 20(4), 481–494 (2006)Google Scholar
  3. 3.
    Brunst, H., Hackenberg, D., Juckeland, G., Rohling, H.: Comprehensive performance tracking with vampir 7. In: M.S. Müller, M.M. Resch, A. Schulz, W.E. Nagel (eds.) Tools for High Performance Computing 2009, pp. 17–29. Springer Berlin Heidelberg (2010). DOI http://dx.doi.org/10.1007/978-3-642-11261-4_2
  4. 4.
    Coulomb, K., Faverge, M., Jazeix, J., Lagrasse, O., Marcoueille, J., Noisette, P., Redondy, A., Vuchener, C.: Visual trace explorer (vite) (2009)Google Scholar
  5. 5.
    Dongarra, J., Meuer, H., Strohmaier, E.: Top500 supercomputer sites. Supercomputer 13, 89–111 (1997)Google Scholar
  6. 6.
    Gmbh, G.T.: Vampir 7 User Manual. Technische Universität Dresden, Blasewitzer Str. 43, 01307 Dresden, Germany, 2011-11-11 / vampir 7.5 edn. (2011)Google Scholar
  7. 7.
    Heath, M., Etheridge, J.: Visualizing the performance of parallel programs. IEEE software 8(5), 29–39 (1991)CrossRefGoogle Scholar
  8. 8.
    Ihaka, R., Gentleman, R.: R: A language for data analysis and graphics. Journal of computational and graphical statistics pp. 299–314 (1996)Google Scholar
  9. 9.
    Johnson, B., Shneiderman, B.: Tree-maps: a space-filling approach to the visualization of hierarchical information structures. In: Proceedings of the IEEE Conference on Visualization, pp. 284–291. IEEE Computer Society Press Los Alamitos, CA, USA (1991). DOI 10.1109/VISUAL.1991.175815Google Scholar
  10. 10.
    Joshi, A., Phansalkar, A., Eeckhout, L., John, L.K.: Measuring benchmark similarity using inherent program characteristics. IEEE Transactions on Computers 55, 769–782 (2006). DOI http://doi.ieeecomputersociety.org/10.1109/TC.2006.85 Google Scholar
  11. 11.
    Kalé, L.V., Zheng, G., Lee, C.W., Kumar, S.: Scaling applications to massively parallel machines using projections performance analysis tool. Future Generation Comp. Syst. 22(3), 347–358 (2006)CrossRefGoogle Scholar
  12. 12.
    de Kergommeaux, J.C., de Oliveira Stein, B., Bernard, P.E.: Pajé, an interactive visualization tool for tuning multi-threaded parallel applications. Parallel Computing 26(10), 1253–1274 (2000)MATHCrossRefGoogle Scholar
  13. 13.
    Knupfer, A., Nagel, W.: Construction and compression of complete call graphs for post-mortem program trace analysis. In: Parallel Processing, 2005. ICPP 2005. International Conference on, pp. 165–172 (2005). DOI 10.1109/ICPP.2005.28Google Scholar
  14. 14.
    Lee, C., Mendes, C., Kalé, L.: Towards scalable performance analysis and visualization through data reduction. In: IEEE International Symposium on Parallel and Distributed Processing (IPDPS), pp. 1–8. IEEE (2008). DOI http://dx.doi.org/10.1109/IPDPS.2008.4536187
  15. 15.
    Lubeck, O., Lang, M., Srinivasan, R., Johnson, G.: Implementation and performance modeling of deterministic particle transport (sweep3d) on the ibm cell/b.e. Scientific Programming 17 (2009)Google Scholar
  16. 16.
    Mohror, K., Karavanic, K.L.: Evaluating similarity-based trace reduction techniques for scalable performance analysis. In: Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC ’09, pp. 55:1–55:12. ACM, New York, NY, USA (2009). DOI http://doi.acm.org/10.1145/1654059.1654115. URL http://doi.acm.org/10.1145/1654059.1654115
  17. 17.
    Nickolayev, O., Roth, P., Reed, D.: Real-time statistical clustering for event trace reduction. International Journal of High Performance Computing Applications 11(2), 144 (1997)CrossRefGoogle Scholar
  18. 18.
    Pillet, V., Labarta, J., Cortes, T., Girona, S.: Paraver: A tool to visualise and analyze parallel code. In: Proceedings of Transputer and occam Developments, WOTUG-18., Transputer and Occam Engineering, vol. 44, pp. 17–31. [S.l.]: IOS Press, Amsterdam (1995)Google Scholar
  19. 19.
    Schnorr, L.M., Huard, G., Navaux, P.O.A.: A hierarchical aggregation model to achieve visualization scalability in the analysis of parallel applications. Parallel Computing 38(3), 91–110 (2012). DOI 10.1016/j.parco.2011.12.001CrossRefGoogle Scholar
  20. 20.
    Schnorr, L.M., Legrand, A., Vincent, J.M.: Detection and analysis of resource usage anomalies in large distributed systems through multi-scale visualization. Concurrency and Computation: Practice and Experience 24(15), 1792–1816 (2012). DOI 10.1002/cpe.1885CrossRefGoogle Scholar
  21. 21.
    Schnorr, L.M., de Oliveira Stein, B., de Kergommeaux, J., Mounié, G.: Pajé trace file format. Tech. rep., ID-IMAG, Grenoble, France (2012). http://paje.sf.net
  22. 22.
    Shende, S., Malony, A.: The tau parallel performance system. International Journal of High Performance Computing Applications 20(2), 287 (2006)CrossRefGoogle Scholar
  23. 23.
    Wickham, H.: ggplot2: elegant graphics for data analysis. Springer-Verlag New York Inc (2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.INRIA MESCAL Research Team, CNRS LIG LaboratoryGrenobleFrance

Personalised recommendations