Transparent offloading of computational hotspots from binary code to Xeon Phi

M. Damschen, H. Riebler, G.F. Vaz, C. Plessl, in:, Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE), EDA Consortium / IEEE, 2015, pp. 1078–1083.

Download
Restricted 238-plessl15_date.pdf 380.55 KB
Conference Paper
Abstract
In this paper, we study how binary applications can be transparently accelerated with novel heterogeneous computing resources without requiring any manual porting or developer-provided hints. Our work is based on Binary Acceleration At Runtime (BAAR), our previously introduced binary acceleration mechanism that uses the LLVM Compiler Infrastructure. BAAR is designed as a client-server architecture. The client runs the program to be accelerated in an environment, which allows program analysis and profiling and identifies and extracts suitable program parts to be offloaded. The server compiles and optimizes these offloaded program parts for the accelerator and offers access to these functions to the client with a remote procedure call (RPC) interface. Our previous work proved the feasibility of our approach, but also showed that communication time and overheads limit the granularity of functions that can be meaningfully offloaded. In this work, we motivate the importance of a lightweight, high-performance communication between server and client and present a communication mechanism based on the Message Passing Interface (MPI). We evaluate our approach by using an Intel Xeon Phi 5110P as the acceleration target and show that the communication overhead can be reduced from 40% to 10%, thus enabling even small hotspots to benefit from offloading to an accelerator.
Publishing Year
Proceedings Title
Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE)
Page
1078-1083
LibreCat-ID

Cite this

Damschen M, Riebler H, Vaz GF, Plessl C. Transparent offloading of computational hotspots from binary code to Xeon Phi. In: Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE). EDA Consortium / IEEE; 2015:1078-1083. doi:10.7873/DATE.2015.1124.
Damschen, M., Riebler, H., Vaz, G. F., & Plessl, C. (2015). Transparent offloading of computational hotspots from binary code to Xeon Phi. In Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE) (pp. 1078–1083). EDA Consortium / IEEE. http://doi.org/10.7873/DATE.2015.1124
@inbook{Damschen_Riebler_Vaz_Plessl_2015, title={Transparent offloading of computational hotspots from binary code to Xeon Phi}, DOI={10.7873/DATE.2015.1124}, booktitle={Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE)}, publisher={EDA Consortium / IEEE}, author={Damschen, Marvin and Riebler, Heinrich and Vaz, Gavin Francis and Plessl, Christian}, year={2015}, pages={1078–1083}}
Damschen, Marvin, Heinrich Riebler, Gavin Francis Vaz, and Christian Plessl. “Transparent Offloading of Computational Hotspots from Binary Code to Xeon Phi.” In Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE), 1078–83. EDA Consortium / IEEE, 2015. doi:10.7873/DATE.2015.1124.
M. Damschen, H. Riebler, G. F. Vaz, and C. Plessl, “Transparent offloading of computational hotspots from binary code to Xeon Phi,” in Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE), 2015, pp. 1078–1083.
Damschen, Marvin et al. “Transparent Offloading of Computational Hotspots from Binary Code to Xeon Phi.” Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE). EDA Consortium / IEEE, 2015. 1078–1083. Web.
Main File(s)
File Name
238-plessl15_date.pdf 380.55 KB
Access Level
Restricted Closed Access
Last Uploaded
2018-03-21T10:29:49Z


Export

Marked Publications

Open Data LibreCat

Search this title in

Google Scholar