Language-Agnostic Optimization and Parallelization for Interpreted Languages
Scientists are increasingly turning to interpreted languages, such as Python, Java, R, Matlab, and Perl, to implement their data analysis algorithms. While such languages permit rapid software development, their implementations often run into performance issues that slow down the scientific process. Source-level approaches for parallelization are problematic for two reasons: first, many of the language features common to these languages can be challenging for the kinds of analyses needed for parallelization; and second, even where such analysis is possible, a language-specific approach implies that each language would need its own parallelizing compiler and/or constructs, resulting in significant duplication of effort.
The Science Up To Par project is investigating a radically different approach to this problem: automatic parallelization at the machine code level using trace information. The key to accomplishing this will be the static and dynamic analysis of executables and the reconstitution of such executables into parallel executables. The key insight is that with trace information it should be possible optimize out the interpreter and other dynamic features in a language-agnostic manner and create parallelized executables for multicore architectures. If successful, this can enable scientists to continue to develop in programming environments that most conveniently support their scientific exploration without paying the performance overheads currently associated with many such environments.
- 1.Bolz, C.F., Cuni, A., Fijalkowski, M., Rigo, A.: Tracing the meta-level: Pypy’s tracing JIT compiler. In: Proceedings of the 4th Workshop on the Implementation, Compilation, Optimization of Object-Oriented Languages and Programming Systems, pp. 18–25. ACM (2009)Google Scholar
- 2.Catanzaro, B., et al.: SEJITS: getting productivity and performance with selective embedded JIT specialization. Technical report UCB/EECS-2010-23, EECS Department, University of California, Berkeley, March 2010Google Scholar
- 3.Danford, F., Welch, E., Cárdenas-Ródriguez, J., Strout, M.M.: Analyzing parallel programming models for magnetic resonance imaging. In: Ding, C., Criswell, J., Wu, P. (eds.) LCPC 2016. LNCS, vol. 10136, pp. 188–202. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-52709-3_15CrossRefGoogle Scholar
- 4.Gaska, B.J.: Parforpy: loop parallelism in python. Master’s thesis, University of Arizona (2017)Google Scholar
- 5.Gaska, B.J., Jothi, N., Mohammadi, M.S., Volk, K., Strout, M.M.: Handling nested parallelism, load imbalance, and early termination in an orbital analysis code. Technical report arXiv:1707.09668, University of Arizona (2017)
- 6.Kotzmann, T., Wimmer, C., Mössenböck, H., Rodriguez, T., Russell, K., Cox, D.: Design of the Java hotspot™ client compiler for Java 6. ACM Trans. Archit. Code Optim. 5(1), 7:1–7:32 (2008)Google Scholar
- 7.Lindenbaum, P.: Programming language use distribution from recent programs/articles, April 2017. https://www.biostars.org/p/251002/
- 8.Oh, T., Beard, S.R., Johnson, N.P., Popovych, S., August, D.I.: A generalized framework for automatic scripting language parallelization. In: Proceedings of the 26th International Conference on Parallel Architectures and Compilation Techniques (PACT) (2017, to appear)Google Scholar
- 9.Oh, T., Kim, H., Johnson, N.P., Lee, J.W., August, D.I.: Practical automatic loop specialization. In: Proceedings of the Eighteenth International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2013, pp. 419–430. ACM, New York (2013)Google Scholar
- 10.Schwartz, E.J., Avgerinos, T., Brumley, D.: All you ever wanted to know about dynamic taint analysis and forward symbolic execution (but might have been afraid to ask). In: Proceedings of IEEE Symposium on Security and Privacy, pp. 317–331 (2010)Google Scholar
- 11.Sharif, M., Lanzi, A., Giffin, J., Lee, W.: Automatic reverse engineering of malware emulators. In: 2009 30th IEEE Symposium on Security and Privacy, pp. 94–109. IEEE (2009)Google Scholar
- 12.Yadegari, B., Debray, S.: Bit-level taint analysis. In: IEEE International Working Conference on Source Code Analysis and Manipulation (SCAM) (2014)Google Scholar
- 13.Yadegari, B., Debray, S.: Symbolic execution of obfuscated code. In: Proceedings of 22nd ACM Conference on Computer and Communications Security (CCS), October 2015Google Scholar