Abstract
OpenMP 5.0 introduced the
directive to support compile-time selection from a set of directive variants based on OpenMP context. OpenMP 5.1 extended context information to include user-defined conditions that enable user-guided runtime adaptation. However, defining conditions that capture the complex interactions between applications and hardware platforms to select an optimized variant is challenging for programmers. This paper explores a novel approach to automate runtime adaptation through machine learning. We design a new
directive to describe semantics for model-driven adaptation and also develop a prototype implementation. Using the Smith-Waterman algorithm as a use-case, our experiments demonstrate that the proposed adaptive OpenMP extension automatically chooses the code variants that deliver the best performance in heterogeneous platforms that consist of CPU and GPU processing capabilities. Using decision tree models for tuning has an accuracy of up to 93.1% in selecting the optimal variant, with negligible runtime overhead.
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
Ansel, J., et al.: OpenTuner: an extensible framework for program autotuning. In: Proceedings of the 23rd International Conference on Parallel Architectures and Compilation, pp. 303–316 (2014)
Ashouri, A.H., Killian, W., Cavazos, J., Palermo, G., Silvano, C.: A survey on compiler autotuning using machine learning. ACM Comput. Surv. (CSUR) 51(5), 1–42 (2018)
Bari, M.A.S., et al.: ARCS: adaptive runtime configuration selection for power-constrained OpenMP applications. In: 2016 IEEE International Conference on Cluster Computing (CLUSTER), pp. 461–470, September 2016. https://doi.org/10.1109/CLUSTER.2016.39
Beckingsale, D., Pearce, O., Laguna, I., Gamblin, T.: Apollo: reusable models for fast, dynamic tuning of input-dependent code. In: 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS), pp. 307–316. IEEE (2017)
Chen, C., Chame, J., Hall, M.: CHiLL: a framework for composing high-level loop transformations. Technical report, Citeseer (2008)
Chen, R.S., Hollingsworth, J.K.: ANGEL: a hierarchical approach to multi-objective online auto-tuning. In: Proceedings of the 5th International Workshop on Runtime and Operating Systems for Supercomputers, pp. 1–8 (2015)
Culjak, I., Abram, D., Pribanic, T., Dzapo, H., Cifrek, M.: A brief introduction to OpenCV. In: 2012 Proceedings of the 35th International Convention MIPRO, pp. 1725–1730. IEEE (2012)
Cummins, C., Petoumenos, P., Wang, Z., Leather, H.: End-to-end deep learning of optimization heuristics. In: 2017 26th International Conference on Parallel Architectures and Compilation Techniques (PACT), pp. 219–232. IEEE (2017)
Fursin, G., et al.: Milepost GCC: machine learning enabled self-tuning compiler. Int. J. Parallel Prog. 39(3), 296–327 (2011)
Grewe, D., Wang, Z., O’Boyle, M.F.: Portable mapping of data parallel programs to OpenCL for heterogeneous systems. In: Proceedings of the 2013 IEEE/ACM International Symposium on Code Generation and Optimization (CGO), pp. 1–10. IEEE (2013)
Hayashi, A., Ishizaki, K., Koblents, G., Sarkar, V.: Machine-learning-based performance heuristics for runtime CPU/GPU selection. In: Proceedings of the Principles and Practices of Programming on the Java Platform, pp. 27–36 (2015)
Hornung, R.D., Keasler, J.A.: The RAJA portability layer: overview and status (2014)
Liao, C., Quinlan, D.J., Panas, T., de Supinski, B.R.: A ROSE-based OpenMP 3.0 research compiler supporting multiple runtime libraries. In: Sato, M., Hanawa, T., Müller, M.S., Chapman, B.M., de Supinski, B.R. (eds.) IWOMP 2010. LNCS, vol. 6132, pp. 15–28. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-13217-9_2
Liao, C., Quinlan, D.J., Vuduc, R., Panas, T.: Effective source-to-source outlining to support whole program empirical optimization. In: Gao, G.R., Pollock, L.L., Cavazos, J., Li, X. (eds.) LCPC 2009. LNCS, vol. 5898, pp. 308–322. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-13374-9_21
Liao, C., Yan, Y., de Supinski, B.R., Quinlan, D.J., Chapman, B.: Early experiences with the OpenMP accelerator model. In: Rendell, A.P., Chapman, B.M., Müller, M.S. (eds.) IWOMP 2013. LNCS, vol. 8122, pp. 84–98. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40698-0_7
Luk, C.K., Hong, S., Kim, H.: Qilin: exploiting parallelism on heterogeneous multiprocessors with adaptive mapping. In: 2009 42nd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO), pp. 45–55. IEEE (2009)
Nugteren, C., Codreanu, V.: CLTune: a generic auto-tuner for OpenCL kernels. In: 2015 IEEE 9th International Symposium on Embedded Multicore/Many-Core Systems-on-Chip, pp. 195–202. IEEE (2015)
OpenMP Architecture Review Board: OpenMP Application Programming Interface 5.0, November 2018. https://www.openmp.org/wp-content/uploads/OpenMP-API-Specification-5.0.pdf
OpenMP Architecture Review Board: OpenMP Application Programming Interface 5.1, November 2020. https://www.openmp.org/wp-content/uploads/OpenMP-API-Specification-5-1.pdf
Pennycook, S.J., Sewall, J.D., Hammond, J.R.: Evaluating the impact of proposed OpenMP 5.0 features on performance, portability and productivity. In: 2018 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC), pp. 37–46, November 2018. https://doi.org/10.1109/P3HPC.2018.00007
Quinlan, D., Liao, C.: The ROSE source-to-source compiler infrastructure. In: Cetus Users and Compiler Infrastructure Workshop, in Conjunction with PACT, vol. 2011, p. 1. Citeseer (2011)
Smith, T.F., Waterman, M.S., et al.: Identification of common molecular subsequences. J. Mol. Biol. 147(1), 195–197 (1981)
Sreenivasan, V., Javali, R., Hall, M., Balaprakash, P., Scogland, T.R.W., de Supinski, B.R.: A framework for enabling OpenMP autotuning. In: Fan, X., de Supinski, B.R., Sinnen, O., Giacaman, N. (eds.) IWOMP 2019. LNCS, vol. 11718, pp. 50–60. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-28596-8_4
Tapus, C., Chung, I.H., Hollingsworth, J.K.: Active harmony: towards automated performance tuning. In: Proceedings of the 2002 ACM/IEEE Conference on Supercomputing, SC 2002, pp. 1–11. IEEE Computer Society Press, Washington, DC (2002)
Wang, Z., O’Boyle, M.: Machine learning in compiler optimization. Proc. IEEE 106(11), 1879–1901 (2018)
Wood, C., et al.: Artemis: automatic runtime tuning of parallel execution parameters using machine learning. In: Chamberlain, B.L., Varbanescu, A.-L., Ltaief, H., Luszczek, P. (eds.) ISC High Performance 2021. LNCS, vol. 12728, pp. 453–472. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-78713-4_24
Yan, Y., Wang, A., Liao, C., Scogland, T.R.W., de Supinski, B.R.: Extending OpenMP Metadirective semantics for runtime adaptation. In: Fan, X., de Supinski, B.R., Sinnen, O., Giacaman, N. (eds.) IWOMP 2019. LNCS, vol. 11718, pp. 201–214. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-28596-8_14
Yi, Q., Seymour, K., You, H., Vuduc, R., Quinlan, D.: POET: parameterized optimizations for empirical tuning. In: 2007 IEEE International Parallel and Distributed Processing Symposium, pp. 1–8. IEEE (2007)
Acknowledgment
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 (LLNL-CONF-826432). The OpenMP language extension work was supported by the U.S. Dept. of Energy, Office of Science, Advanced Scientific Computing Research. The compiler and runtime work were supported by LLNL-LDRD 21-ERD-018.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Artifact Availability Statement
Artifact Availability Statement
Summary of the Experiments Reported: We ran Smith Waterman algorithm on LLNL’s Corona and Pascal supercomputer. The detailed software configurations are given in the experiment section of the paper.
Software Artifact Availability: All author-created software artifacts are maintained in a public repository under an OSI-approved license.
Hardware Artifact Availability: There are no author-created hardware artifacts.
Data Artifact Availability: All author-created data artifacts are maintained in a public repository under an OSI-approved license.
Proprietary Artifacts: None of the associated artifacts, author-created or otherwise, are proprietary.
List of URLs and/or DOIs where artifacts are available:
Rights and permissions
Copyright information
© 2022 Anjia Wang and Yonghong Yan, and Lawrence Livermore National Security, LLC, under exclusive license to Springer Nature Switzerland AG, part of Springer Nature
About this paper
Cite this paper
Liao, C. et al. (2022). Extending OpenMP for Machine Learning-Driven Adaptation. In: Bhalachandra, S., Daley, C., Melesse Vergara, V. (eds) Accelerator Programming Using Directives. WACCPD 2021. Lecture Notes in Computer Science(), vol 13194. Springer, Cham. https://doi.org/10.1007/978-3-030-97759-7_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-97759-7_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-97758-0
Online ISBN: 978-3-030-97759-7
eBook Packages: Computer ScienceComputer Science (R0)