Abstract
The performance of collective operations has been a critical issue since the advent of Message Passing Interface (MPI). Many algorithms have been proposed for each MPI collective operation but none of them proved optimal in all situations. Different algorithms demonstrate superior performance depending on the platform, the message size, the number of processes, etc. MPI implementations perform the selection of the collective algorithm empirically, executing a simple runtime decision function. While efficient, this approach does not guarantee the optimal selection. As a more accurate but equally efficient alternative, the use of analytical performance models of collective algorithms for the selection process was proposed and studied. Unfortunately, the previous attempts in this direction have not been successful.
We revisit the analytical model-based approach and propose two innovations that significantly improve the selective accuracy of analytical models: (1) We derive analytical models from the code implementing the algorithms rather than from their high-level mathematical definitions. This results in more detailed models. (2) We estimate model parameters separately for each collective algorithm and include the execution of this algorithm in the corresponding communication experiment.
We experimentally demonstrate the accuracy and efficiency of our approach using Open MPI broadcast algorithms and two different Grid’5000 clusters.
This publication has emanated from research conducted with the financial support of Science Foundation Ireland (SFI) under Grant Number 14/IA/2474.
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
A Message-Passing Interface Standard. https://www.mpi-forum.org/. Accessed 8 Mar 2021
Rabenseifner, R.: Automatic profiling of MPI applications with hardware performance counters. In: Dongarra, J., Luque, E., Margalef, T. (eds.) EuroPVM/MPI 1999. LNCS, vol. 1697, pp. 35–42. Springer, Heidelberg (1999). https://doi.org/10.1007/3-540-48158-3_5
Open MPI: Open Source High Performance Computing. https://www.open-mpi.org/. Accessed 8 Mar 2021
MPICH - A Portable Implementation of MPI. http://www.mpich.org/. Accessed 8 Mar 2021
Thakur, R., Rabenseifner, R., Gropp, W.: Optimization of collective communication operations in MPICH. Int. J. High Perform. Comput. Appl. 19(1), 49–66 (2005)
Gabriel, E., et al.: Open MPI: goals, concept, and design of a next generation MPI implementation. In: Kranzlmüller, D., Kacsuk, P., Dongarra, J. (eds.) EuroPVM/MPI 2004. LNCS, vol. 3241, pp. 97–104. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-30218-6_19
Fagg, G.E., Pjesivac-Grbovic, J., Bosilca, G., Angskun, T., Dongarra, J., Jeannot, E.: Flexible collective communication tuning architecture applied to Open MPI. In: Euro PVM/MPI (2006)
Pješivac-Grbović, J., Angskun, T., Bosilca, G., Fagg, G.E., Gabriel, E., Dongarra, J.J.: Performance analysis of MPI collective operations. Clust. Comput. 10(2), 127–143 (2007)
Hockney, R.W.: The communication challenge for MPP: Intel Paragon and Meiko CS-2. Parallel Comput. 20(3), 389–398 (1994)
Chan, E.W., Heimlich, M.F., Purkayastha, A., van de Geijn, R.A.: On optimizing collective communication. In: IEEE International Conference on Cluster Computing 2004, pp. 145–155 (2004)
Chan, E., Heimlich, M., Purkayastha, A., van de Geijn, R.: Collective communication: theory, practice, and experience: research articles. Concurr. Comput. Pract. Exper. 19(13), 1749–1783 (2007)
Culler, D., Liu, L.T., Martin, R.P., Yoshikawa, C.: LogP performance assessment of fast network interfaces. IEEE Micro 16(1), 35–43 (1996)
Kielmann, T., Bal, H.E., Verstoep, K.: Fast measurement of LogP parameters for message passing platforms. In: Rolim, J. (ed.) IPDPS 2000. LNCS, vol. 1800, pp. 1176–1183. Springer, Heidelberg (2000). https://doi.org/10.1007/3-540-45591-4_162
Lastovetsky, A., Rychkov, V.: Building the communication performance model of heterogeneous clusters based on a switched network. In: IEEE International Conference on Cluster Computing 2007, pp. 568–575 (2007)
Lastovetsky, A., Rychkov, V.: Accurate and efficient estimation of parameters of heterogeneous communication performance models. Int. J. High Perform. Comput. Appl. 23(2), 123–139 (2009)
Lastovetsky, A., Rychkov, V., O’Flynn, M.: Accurate heterogeneous communication models and a software tool for their efficient estimation. Int. J. High Perform. Comput. Appl. 24(1), 34–48 (2010)
Rico-Gallego, J.A., Díaz-Martín, J.C., Manumachu, R.R., Lastovetsky, A.L.: A survey of communication performance models for high-performance computing. ACM Comput. Surv. 51(6), 1–36 (2019)
Pješivac–Grbović, J., Fagg, G.E., Angskun, T., Bosilca, G., Dongarra, J.J.: MPI collective algorithm selection and quadtree encoding. In: Mohr, B., Träff, J.L., Worringen, J., Dongarra, J. (eds.) EuroPVM/MPI 2006. LNCS, vol. 4192, pp. 40–48. Springer, Heidelberg (2006). https://doi.org/10.1007/11846802_14
Pješivac-Grbović, J., Bosilca, G., Fagg, G.E., Angskun, T., Dongarra, J.J.: Decision trees and MPI collective algorithm selection problem. In: Kermarrec, A.-M., Bougé, L., Priol, T. (eds.) Euro-Par 2007. LNCS, vol. 4641, pp. 107–117. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-74466-5_13
Quinlan, J.R.: C4.5: Programs for Machine Learning. Morgan Kaufmann Publishers Inc., Burlington (1993)
Hunold, S., Bhatele, A., Bosilca, G., Knees, P.: Predicting MPI collective communication performance using machine learning. In: IEEE International Conference on Cluster Computing 2020, pp. 259–269 (2020)
Wickramasinghe, U., Lumsdaine, A.: A survey of methods for collective communication optimization and tuning. arXiv preprint arXiv:1611.06334 (2016)
Grid5000. http://www.grid5000.fr. Accessed 8 Mar 2021
Lastovetsky, A., Rychkov, V., O’Flynn, M.: MPIBlib: benchmarking MPI communications for parallel computing on homogeneous and heterogeneous clusters. In: Lastovetsky, A., Kechadi, T., Dongarra, J. (eds.) EuroPVM/MPI 2008. LNCS, vol. 5205, pp. 227–238. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-87475-1_32
Huber, P.J.: Robust estimation of a location parameter. In: Kotz, S., Johnson, N.L. (eds.) Breakthroughs in Statistics. Springer Series in Statistics (Perspectives in Statistics), pp. 492–518. Springer, New York (1992). https://doi.org/10.1007/978-1-4612-4380-9_35
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Nuriyev, E., Lastovetsky, A. (2021). A New Model-Based Approach to Performance Comparison of MPI Collective Algorithms. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 2021. Lecture Notes in Computer Science(), vol 12942. Springer, Cham. https://doi.org/10.1007/978-3-030-86359-3_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-86359-3_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86358-6
Online ISBN: 978-3-030-86359-3
eBook Packages: Computer ScienceComputer Science (R0)