Peng Jiang
ABSTRACT
Finite State Machine (FSM) is the key kernel behind many popular applications, including regular expression matching, text tokenization, and Huffman decoding. Parallelizing FSMs is extremely difficult because of the strong dependencies and unpredictable memory accesses. Previous efforts have largely focused on multi-core parallelization, and used different approaches, including {\em speculative} and {\em enumerative} execution, both of which have been effective but also have limitations. With increasing width and improving flexibility in SIMD instruction sets, this paper focuses on combining SIMD and multi/many-core parallelism for FSMs. We have developed a novel strategy, called {\em enumerative speculation}. Instead of speculating on a single state as in speculative execution or enumerating all possible states as in enumerative execution, our strategy speculates transitions from several possible states, reducing the prediction overheads of speculation approach and the large amount of redundant work in the enumerative approach. A simple lookback approach produces a set of guessed states to achieve high speculation success rates in our enumerative speculation. We evaluate our method with four popular FSM applications: Huffman decoding, regular expression matching, HTML tokenization, and Div7. We obtain up to 2.5x speedup using SIMD on one core and up to 95x combining SIMD with 60 cores of an Intel Xeon Phi. On a single core, we outperform the best single-state speculative execution version by an average of 1.6x, and in combining SIMD and many-core parallelism, outperform enumerative execution by an average of 2x.
- K. Asanovic, R. Bodik, J. Demmel, T. Keaveny, K. Keutzer, J. Kubiatowicz, N. Morgan, D. Patterson, K. Sen, J. Wawrzynek, D. Wessel, and K. Yelick. A view of the parallel computing landscape. Commun. ACM, 52(10):56--67, Oct. 2009. Google Scholar
Digital Library
- G. E. Blelloch. Prefix Sums and Their Applications. Technical Report Carnegie Mellon University-CS-90--190, School of Computer Science, Carnegie Mellon University, Nov. 1990.Google Scholar
- L. Chen, P. Jiang, and G. Agrawal. Exploiting Recent SIMD Architectural Advances for Irregular Applications. In Proceedings of the 2016 International Symposium on Code Generation and Optimization, CGO 2016, 2016. Google ScholarDigital Library
- F. Franchetti and M. Puschel. A SIMD Vectorizing Compiler for Digital Signal Processing Algorithms. In Proceedings of the 2002 IEEE International Parallel and Distributed Processing Symposium (IPDPS 2002), pages 20 --26. IEEE, 2002. Google ScholarCross Ref
- C. Garca, R. Lario, M. Prieto, L. Piuel, and F. Tirado. Vectorization of Multigrid Codes Using SIMD ISA Extensions. Proceedings of the 2003 IEEE Inernational Parallel and Distributed Processing Symposium (IPDPS 2003), 0:58a, 2003.Google ScholarDigital Library
- W. D. Hillis and G. L. Steele, Jr. Data Parallel Algorithms. Commun. ACM, 29(12), Dec. 1986. Google ScholarDigital Library
- J. Holub and S.vStekr. On Parallel Implementations of Deterministic Finite Automata. In Proceedings of the 14th International Conference on Implementation and Application of Automata, CIAA '09, 2009. Google ScholarDigital Library
- N. Ide, M. Hirano, Y. Endo, S. Yoshioka, H. Murakami, A. Kunimatsu, T. Sato, T. Kamei, T. Okada, and M. Suzuoki. 2.44-GFLOPS 300-MHz Floating-Point Vector-Processing Unit for High-Performance 3D Graphics Computing. IEEE Journal of Solid-State Circuits, 35(7):1025 --1033, Jul 2000. Google ScholarCross Ref
- P. Jiang, L. Chen, and G. Agrawal. Reusing data reorganization for efficient simd parallelization of adaptive irregular applications. In Proceedings of the 2016 International Conference on Supercomputing, ICS '16, 2016. Google ScholarDigital Library
- F. Khorasani, K. Vora, R. Gupta, and L. N. Bhuyan. CuSha: Vertex-centric Graph Processing on GPUs. In Proceedings of the 23rd International Symposium on High-performance Parallel and Distributed Computing, HPDC '14, 2014. Google ScholarDigital Library
- S. Kim and H. Han. Efficient simd code generation for irregular kernels. In Proceedings of the 17th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP '12, pages 55--64, New York, NY, USA, 2012. ACM. Google ScholarDigital Library
- S. T. Klein and Y. Wiseman. Parallel Huffman Decoding with Applications to JPEG Files. THE COMPUTER JOURNAL, 46:487--497, 2003. Google ScholarCross Ref
- R. E. Ladner and M. J. Fischer. Parallel prefix computation. J. ACM, 27(4), Oct. 1980. Google ScholarDigital Library
- X. Liu, M. Smelyanskiy, E. Chow, and P. Dubey. Efficient Sparse Matrix-vector Multiplication on x86-based Many-core Processors. ICS '13, 2013.Google ScholarDigital Library
- D. Luchaup, R. Smith, C. Estan, and S. Jha. Speculative Parallel Pattern Matching. IEEE Transactions on Information Forensics and Security, 6(2):438--451, June 2011. Google ScholarDigital Library
- D. Merrill, M. Garland, and A. Grimshaw. Scalable gpu graph traversal. In Proceedings of the 17th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP '12, pages 117--128, New York, NY, USA, 2012. ACM. Google ScholarDigital Library
- T. Mytkowicz, M. Musuvathi, and W. Schulte. Data-parallel finite-state machines. In Proceedings of the 19th International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS '14, pages 529--542, New York, NY, USA, 2014. ACM. Google ScholarDigital Library
- Y. Pan, Y. Zhang, and K. Chiu. Simultaneous Transducers for Data-Parallel XML Parsing. In Parallel and Distributed Processing, 2008. IPDPS 2008. IEEE International Symposium on, pages 1--12, 2008.Google Scholar
- S. J. Pennycook, C. J. Hughes, M. Smelyanskiy, and S. A. Jarvis. Exploring simd for molecular dynamics, using intel xeon processors and intel xeon phi coprocessors. In Proceedings of the 2013 IEEE 27th International Symposium on Parallel and Distributed Processing, IPDPS '13, 2013.Google ScholarDigital Library
- P. Prabhu, G. Ramalingam, and K. Vaswani. Safe Programmable Speculative Parallelism. In Proceedings of Programming Language Design and Implementation (PLDI). Association for Computing Machinery, Inc., June 2010. Google ScholarDigital Library
- J. Qiu, Z. Zhao, and B. Ren. Microspec: Speculation-centric fine-grained parallelization for fsm computations. In Proceedings of the 2016 International Conference on Parallel Architectures and Compilation, PACT '16, pages 221--233, New York, NY, USA, 2016. ACM. Google ScholarDigital Library
- T. Rognes and E. Seeberg. Six-Fold Speed-up of Smithwaterman Sequence Database Searches Using Parallel Processing on Common Microprocessors. 16(8):699--706, 2000.Google Scholar
- E. Saule and U. V. Catalyurek. An early evaluation of the scalability of graph algorithms on the intel mic architecture. In Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum, IPDPSW '12, pages 1629--1639, Washington, DC, USA, 2012. IEEE Computer Society. Google ScholarDigital Library
- W. T. Tang, R. Zhao, M. Lu, Y. Liang, H. P. Huynh, X. Li, and R. S. M. Goh. Optimizing and Auto-tuning Scale-free Sparse Matrix-vector Multiplication on Intel Xeon Phi. In Proceedings of the 13th Annual IEEE/ACM International Symposium on Code Generation and Optimization, CGO '15, 2015. Google ScholarCross Ref
- Z. Zhao and X. Shen. On-the-Fly Principled Speculation for FSM Parallelization. SIGARCH Comput. Archit. News, 43(1), Mar. 2015.Google Scholar
- Z. Zhao, B. Wu, and X. Shen. Challenging the "Embarrassingly Sequential": Parallelizing Finite State Machine-based Computations Through Principled Speculation. SIGPLAN Not., 2014.Google ScholarDigital Library
Index Terms
- Combining SIMD and Many/Multi-core Parallelism for Finite State Machines with Enumerative Speculation
Recommendations
Combining SIMD and Many/Multi-core Parallelism for Finite State Machines with Enumerative Speculation
PPoPP '17Finite State Machine (FSM) is the key kernel behind many popular applications, including regular expression matching, text tokenization, and Huffman decoding. Parallelizing FSMs is extremely difficult because of the strong dependencies and unpredictable ...
Combining SIMD and Many/Multi-core Parallelism for Finite-state Machines with Enumerative Speculation
Special Issue on PPoPP 2017 (Part 2) and Regular PapersFinite-state Machine (FSM) is the key kernel behind many popular applications, including regular expression matching, text tokenization, and Huffman decoding. Parallelizing FSMs is extremely difficult because of the strong dependencies and unpredictable ...
Boundary element quadrature schemes for multi- and many-core architectures
In the paper we study the performance of the regularized boundary element quadrature routines implemented in the BEM4I library developed by the authors. Apart from the results obtained on the classical multi-core architecture represented by the Intel ...
Comments