Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores

conference paper Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores yes Marvin Damschen author Christian Plessl author 161530000-0001-5728-9982 27 department 518 department 78 department SFB 901 project SFB 901 - Subprojekt C2 project SFB 901 - Project Area C project Self-Adaptive Virtualisation-Aware High-Performance/Low-Energy Heterogeneous System Architectures project This paper introduces Binary Acceleration At Runtime(BAAR), an easy-to-use on-the-fly binary acceleration mechanismwhich aims to tackle the problem of enabling existentsoftware to automatically utilize accelerators at runtime. BAARis based on the LLVM Compiler Infrastructure and has aclient-server architecture. The client runs the program to beaccelerated in an environment which allows program analysisand profiling. Program parts which are identified as suitable forthe available accelerator are exported and sent to the server.The server optimizes these program parts for the acceleratorand provides RPC execution for the client. The client transformsits program to utilize accelerated execution on the server foroffloaded program parts. We evaluate our work with a proofof-concept implementation of BAAR that uses an Intel XeonPhi 5110P as the acceleration target and performs automaticoffloading, parallelization and vectorization of suitable programparts. The practicality of BAAR for real-world examples is shownbased on a study of stencil codes. Our results show a speedup ofup to 4 without any developer-provided hints and 5.77 withhints over the same code compiled with the Intel Compiler atoptimization level O2 and running on an Intel Xeon E5-2670machine. Based on our insights gained during implementationand evaluation we outline future directions of research, e.g.,offloading more fine-granular program parts than functions, amore sophisticated communication mechanism or introducing onstack-replacement. https://ris.uni-paderborn.de/download/303/1442/303-plessl15_adapt.pdf application/pdfno 2015 eng Proceedings of the 5th International Workshop on Adaptive Self-tuning Computing Systems (ADAPT) 1412.3906 Damschen, Marvin, and Christian Plessl. “Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores.” In Proceedings of the 5th International Workshop on Adaptive Self-Tuning Computing Systems (ADAPT), 2015. M. Damschen and C. Plessl, “Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores,” 2015. Damschen M, Plessl C. Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores. In: Proceedings of the 5th International Workshop on Adaptive Self-Tuning Computing Systems (ADAPT). ; 2015. @inproceedings{Damschen_Plessl_2015, title={Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores}, booktitle={Proceedings of the 5th International Workshop on Adaptive Self-tuning Computing Systems (ADAPT)}, author={Damschen, Marvin and Plessl, Christian}, year={2015} } M. Damschen, C. Plessl, in: Proceedings of the 5th International Workshop on Adaptive Self-Tuning Computing Systems (ADAPT), 2015. Damschen, Marvin, and Christian Plessl. “Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores.” Proceedings of the 5th International Workshop on Adaptive Self-Tuning Computing Systems (ADAPT), 2015. Damschen, M., & Plessl, C. (2015). Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores. Proceedings of the 5th International Workshop on Adaptive Self-Tuning Computing Systems (ADAPT). 3032017-10-17T12:41:51Z2023-09-26T13:29:59Z