Abstract
OpenMP has supported the offload of computations to accelerators such as GPUs since version 4.0. A crucial aspect in OpenMP offloading is to manage the accelerator data environment. Currently, this has to be explicitly programmed by users, which is non-trival and often results in suboptimal performance. The unified memory feature available in recent GPU architectures introduces another option, implicit management. However, our experiments show that it incurs several performance issues, especially under GPU memory oversubscription. In this paper, we propose a compiler and runtime collaborative approach to manage OpenMP GPU data under unified memory. In our framework, the compiler performs data reuse analysis to assist runtime data management. The runtime combines static and dynamic information to make optimized data management decisions. We have implement the proposed technology in the LLVM framework. The evaluation shows our method can achieve significant performance improvement for OpenMP GPU offloading.
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
Similar content being viewed by others
References
OpenACC. http://www.openacc.org
Summit. https://www.olcf.ornl.gov/summit
OpenMP 4.0 specifications (2013). http://www.openmp.org/wp-content/uploads/OpenMP4.0.0.pdf
Agarwal, N., Nellans, D., Stephenson, M., O’Connor, M., Keckler, S.W.: Page placement strategies for GPUs within heterogeneous memory systems. In: ASPLOS 2015, pp. 607–618. ACM, New York (2015)
Antao, S.F., et al.: Offloading support for OpenMP in Clang and LLVM. In: LLVM-HPC 2016, pp. 1–11. IEEE Press, Piscataway (2016)
Che, S., et al.: Rodinia: a benchmark suite for heterogeneous computing. In: 2009 IEEE International Symposium on Workload Characterization, IISWC 2009, pp. 44–54. IEEE (2009)
Cui, X., Scogland, T.R.W., de Supinski, B.R., Feng, W.C.: Directive-based partitioning and pipelining for graphics processing units. In: 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS), pp. 575–584, May 2017
Grinberg, L., Bertolli, C., Haque, R.: Hands on with OpenMP4.5 and unified memory: developing applications for IBM’s hybrid CPU + GPU systems (Part I). In: de Supinski, B.R., Olivier, S.L., Terboven, C., Chapman, B.M., Müller, M.S. (eds.) IWOMP 2017. LNCS, vol. 10468, pp. 3–16. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-65578-9_1
Hahnfeld, J., Cramer, T., Klemm, M., Terboven, C., Müller, M.S.: A pattern for overlapping communication and computation with OpenMP\(^*\) target directives. In: de Supinski, B.R., Olivier, S.L., Terboven, C., Chapman, B.M., Müller, M.S. (eds.) IWOMP 2017. LNCS, vol. 10468, pp. 325–337. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-65578-9_22
Jablin, T.B., Prabhu, P., Jablin, J.A., Johnson, N.P., Beard, S.R., August, D.I.: Automatic CPU-GPU communication management and optimization. In: PLDI 2011, pp. 142–151. ACM, New York (2011)
Jaleel, A., Theobald, K.B., Steely, Jr., S.C., Emer, J.: High performance cache replacement using re-reference interval prediction (RRIP). In: ISCA 2010, pp. 60–71. ACM, New York (2010)
Lattner, C., Adve, V.: LLVM: a compilation framework for lifelong program analysis & transformation. In: CGO 2004, p. 75. IEEE Computer Society, Washington, DC (2004)
Li, L., Tong, D., Xie, Z., Lu, J., Cheng, X.: Optimal bypass monitor for high performance last-level caches. In: PACT 2012, pp. 315–324. ACM, New York (2012)
Mishra, A., Li, L., Kong, M., Finkel, H., Chapman, B.: Benchmarking and evaluating unified memory for OpenMP GPU offloading. In: LLVM-HPC 2017, pp. 6:1–6:10. ACM, New York (2017)
NVIDIA: Compute unified device architecture programming guide (2007)
Olivier, S.L., Hammond, S.D., Duran, A.: Double buffering for MCDRAM on second generation Intel® Xeon Phi™ processors with OpenMP. In: de Supinski, B.R., Olivier, S.L., Terboven, C., Chapman, B.M., Müller, M.S. (eds.) IWOMP 2017. LNCS, vol. 10468, pp. 311–324. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-65578-9_21
Pai, S., Govindarajan, R., Thazhuthaveetil, M.J.: Fast and efficient automatic memory management for GPUs using compiler-assisted runtime coherence scheme. In: PACT 2012, pp. 33–42. ACM, New York (2012)
Qureshi, M.K., Jaleel, A., Patt, Y.N., Steely, S.C., Emer, J.: Adaptive insertion policies for high performance caching. In: ISCA 2007, pp. 381–391. ACM, New York (2007)
Stone, J.E., Gohara, D., Shi, G.: OpenCL: a parallel programming standard for heterogeneous computing systems. Comput. Sci. Eng. 12(3), 66–73 (2010)
Zhao, J., Xie, Y.: Optimizing bandwidth and power of graphics memory with hybrid memory technologies and adaptive data migration. In: ICCAD 2012, pp. 81–87. ACM, New York (2012)
Acknowledgement
This research was supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. Department of Energy Office of Science and the National Nuclear Security Administration.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Li, L., Finkel, H., Kong, M., Chapman, B. (2018). Manage OpenMP GPU Data Environment Under Unified Address Space. In: de Supinski, B., Valero-Lara, P., Martorell, X., Mateo Bellido, S., Labarta, J. (eds) Evolving OpenMP for Evolving Architectures. IWOMP 2018. Lecture Notes in Computer Science(), vol 11128. Springer, Cham. https://doi.org/10.1007/978-3-319-98521-3_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-98521-3_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-98520-6
Online ISBN: 978-3-319-98521-3
eBook Packages: Computer ScienceComputer Science (R0)