Abstract
We present a new approach for efficient process synchronization in parallel discrete event simulation on multicore computers. We aim specifically at simulation of spatially extended stochastic system models where time intervals between successive inter-process events are highly variable and without lower bounds: This includes models governed by the mesoscopic Reaction-Diffusion Master Equation (RDME). A central part of our approach is a mechanism for optimism control, in which each process disseminates accurate information about timestamps of its future outgoing interprocess events to its neighbours. This information gives each process a precise basis for deciding when to pause local processing to reduce the risk of expensive rollbacks caused by future “delayed” incoming events. We apply our approach to a natural parallelization of the Next Subvolume Method (NSM) for simulating systems obeying RDME. Since this natural parallelization does not expose accurate timestamps of future interprocess events, we restructure it to expose such information, resulting in a simulation algorithm called Refined Parallel NSM (Refined PNSM). We have implemented Refined PNSM in a parallel simulator for spatial extended Markovian processes. On 32 cores, it achieves an efficiency ranging between 43--95% for large models, and on average 37% for small models, compared to an efficient sequential simulation without any code for parallelization. It is shown that the gain of restructuring the naive parallelization into Refined PNSM more than outweighs its overhead. We also show that our resulting simulator is superior in performance to existing simulators on multicores for comparable models.
- P. Bauer and S. Engblom. 2015. Sensitivity estimation and inverse problems in spatial stochastic models of chemical kinetics. Lecture Notes in Computational Science and Engineering, Vol. 103. Springer, 519--527.Google Scholar
- P. Bauer and S. Engblom. 2017. The URDME Manual version 1.3. Technical Report 2017-003. Dept of Information Technology, Uppsala University. Retrieved from http://arxiv.org/abs/0902.2912.Google Scholar
- P. Bauer, J. Lindén, S. Engblom, and B. Jonsson. 2015. Efficient inter-process synchronization for parallel discrete event simulation on multicores. In Proceedings of the ACM SIGSIM Conference on Principles of Advanced Discrete Simulation (SIGSIM-PADS’15). ACM, 183--194. Google ScholarDigital Library
- O. G. Berg and P. H. von Hippel. 1985. Diffusion-controlled macromolecular interactions. Ann. Rev. Biophys. Chem. 14, 1 (1985), 131--160.Google ScholarCross Ref
- M. A. Gibson and J. Bruck. 2000. Efficient exact stochastic simulation of chemical systems with many species and many channels. J. Phys. Chem. A 104, 9 (2000), 1876--1889.Google ScholarCross Ref
- C. D. Carothers, D. Bauer, and S. Pearce. 2000. ROSS: A high-performance, low memory, modular time warp system. In Proceedings of the 14th Workshop on Parallel and Distributed Simulation. IEEE, 53--60. Google ScholarDigital Library
- L.-L. Chen, Y.-S. Lu, Y. P. Yao, S. L. Peng, and L.-D. Wu. 2011. A well-balanced time warp system on multi-core environments. In Proceedings of the 25th Workshop on Parallel and Distributed Simulation. IEEE, 1--9. Google ScholarDigital Library
- S. R. Das. 2000. Adaptive protocols for parallel discrete event simulation. J. Oper. Res. Soc. 51, 4 (2000), 385--394.Google ScholarCross Ref
- E. Deelman and B. K. Szymanski. 1997. Breadth-first rollback in spatially explicit simulations. In Proceedings of the 11th Workshop on Parallel and Distributed Simulation. IEEE, 124--131. Google ScholarDigital Library
- L. Dematté and T. Mazza. 2008. On parallel stochastic simulation of diffusive systems. In Proceedings of the 6th Int’l Conference on Computational Methods in Systems Biology (CMSB’08). Springer-Verlag, Berlin, Heidelberg, 191--210. Google ScholarDigital Library
- B. Drawert, S. Engblom, and A. Hellander. 2012. URDME: A modular framework for stochastic simulation of reaction-transport processes in complex geometries. BMC Syst. Biol. 6, 76 (2012), 1--17.Google ScholarCross Ref
- J. Elf and M. Ehrenberg. 2004. Spontaneous separation of bi-stable biochemical systems into spatial domains of opposite phases. Syst. Biol. 1, 2 (2004), 230--236.Google ScholarCross Ref
- S. Engblom. 2008. Numerical Solution Methods in Stochastic Chemical Kinetics. Ph.D. Dissertation. Uppsala University.Google Scholar
- D. Fange and J. Elf. 2006. Noise-induced min phenotypes in E. coli. PLoS Comput. Biol. 2, 6 (2006), e80.Google ScholarCross Ref
- A. Ferscha. 1995. Probabilistic adaptive direct optimism control in time warp. In Proceedings of the 9th Workshop on Parallel and Distributed Simulation. IEEE, 120--129. Google ScholarDigital Library
- R. M. Fujimoto. 1990. Parallel discrete event simulation. Commun. ACM 33, 10 (1990), 30--53. Google ScholarDigital Library
- R. M. Fujimoto. 2016. Research challenges in parallel and distributed simulation. ACM Trans. Model. Comput. Simul. 26, 4 (2016), 22:1--22:29. Google ScholarDigital Library
- C. W. Gardiner. 2007. Handbook of Stochastic Methods for Physics, Chemistry, and the Natural Sciences. Springer.Google Scholar
- D. T. Gillespie. 1977. Exact stochastic simulation of coupled chemical reactions. J. Phys. Chem. 81, 25 (1977), 2340--2361.Google ScholarCross Ref
- D. T. Gillespie. 1992. A rigorous derivation of the chemical master equation. Phys. A 188 (1992), 404--425.Google ScholarCross Ref
- S. Jafer, Q. Liu, and G. A. Wainer. 2013. Synchronization methods in parallel and distributed discrete-event simulation. Simul. Model. Pract. Th. 30 (2013), 54--73.Google ScholarCross Ref
- D. R. Jefferson. 1985. Virtual time. ACM Trans. Program. Lang. Syst. 7, 3 (1985), 404--425. Google ScholarDigital Library
- M. Jeschke, R. Ewald, A. Park, R. Fujimoto, and A. M. Uhrmacher. 2008. A parallel and distributed discrete event approach for spatial cell-biological simulations. SIGMETRICS Perf. Eval. Rev. 35 (2008), 22--31. Google ScholarDigital Library
- G. Karypis and V. Kumar. 1998. A fast and high quality multilevel scheme for partitioning irregular graphs. SIAM J. Sci. Comput. 20, 1 (1998), 359--392. Google ScholarDigital Library
- M. J. Lawson, B. Drawert, M. Khammash, L. Petzold, and T. M. Yi. 2013. Spatial stochastic dynamics enable robust cell polarization. PLoS Comp. Biol. 9, 7 (2013), e1003139.Google ScholarCross Ref
- Z. Lin, C. Tropper, R. A. McDougal, M. N. I. Patoary, W. W. Lytton, Y. Yao, and M. L. Hines. 2017. Multithreaded stochastic PDES for reactions and diffusions in neurons. ACM Trans. Model. Comput. Simul. 27, 2 (2017), 7:1--7:27. Google ScholarDigital Library
- J. Lindén, P. Bauer, S. Engblom, and B. Jonsson. 2017. Exposing inter-process information for efficient parallel discrete event simulation of spatial stochastic systems. In Proceedings of the ACM SIGSIM Conference on Principles of Advanced Discrete Simulation (SIGSIM-PADS’17). ACM, 53--64. Google ScholarDigital Library
- J. Liu and D. Nicol. 2002. Lookahead revisited in wireless network simulations. In Proceedings of the 16th Workshop on Parallel and Distributed Simulation. IEEE, 79--88. Google ScholarDigital Library
- B. Lubachevsky, A. Schwartz, and A. Weiss. 1991. An analysis of rollback-based simulation. ACM Trans. Model. Comput. Simul. 1, 2 (1991), 154--193. Google ScholarDigital Library
- B. D. Lubachevsky. 1988. Efficient parallel simulations of dynamic Ising spin systems. J. Comput. Phys. 75, 1 (1988), 103--122. Google ScholarDigital Library
- N. Marziale, F. Nobilia, A. Pellegrini, and F. Quaglia. 2016. Granular time warp objects. In Proceedings of the ACM SIGSIM Conference on Principles of Advanced Discrete Simulation (SIGSIM-PADS’16). ACM, 57--68. Google ScholarDigital Library
- M. Matsumoto and T. Nishimura. 1998. Mersenne Twister: A 623-dimensionally equidistributed uniform pseudo-random number generator. ACM Trans. Model. Comput. Simul. 8, 1 (Jan. 1998), 3--30. Google ScholarDigital Library
- J. Misra. 1986. Distributed discrete-event simulation. ACM Comput. Surv. 18, 1 (1986), 39--65. Google ScholarDigital Library
- A. Pellegrini and F. Quaglia. 2015. NUMA time warp. In Proceedings of the ACM SIGSIM Conference on Principles of Advanced Discrete Simulation (SIGSIM-PADS’15). ACM, 59--70. Google ScholarDigital Library
- A. Pellegrini, R. Vitali, S. Peluso, and F. Quaglia. 2012. Transparent and efficient shared-state management for optimistic simulations on multi-core machines. In Proceedings of the International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS’12). IEEE, 134--141. Google ScholarDigital Library
- P. L. Reiher, F. Wieland, and D. Jefferson. 1989. Limitation of optimism in the time warp operating system. In Proceedings of the 21st Winter Simulation Conference. ACM, 765--770. Google ScholarDigital Library
- R. B. Schinazi. 1997. Predator-prey and host-parasite spatial stochastic models. Ann. Appl. Probab. 7, 1 (1997), 1--9.Google ScholarCross Ref
- L. M. Sokol, J. B. Weissman, and P. A. Mutchler. 1991. MTW: An empirical performance study. In Proceedings of the 23rd Winter Simulation Conference. IEEE, 557--563. Google ScholarDigital Library
- S. Srinivasan and P. F. Reynolds Jr. 1998. Elastic time. ACM Trans. Model. Comput. Simul. 8, 2 (1998), 103--139. Google ScholarDigital Library
- J. S. Steinman. 1993. Breathing time warp. In Proceedings of the 7th Workshop on Parallel and Distributed Simulation. ACM, 109--118. Google ScholarDigital Library
- M. Sturrock, A. Hellander, A. Matzavinos, and M. A. J. Chaplain. 2013. Spatial stochastic modelling of the Hes1 gene regulatory network: Intrinsic noise can explain heterogeneity in embryonic stem cell differentiation. J. R. Soc. Interface 10, 80 (2013), 20120988.Google ScholarCross Ref
- N. G. van Kampen. 2004. Stochastic Processes in Physics and Chemistry (2nd ed.). Elsevier.Google Scholar
- B. Wang, B. Hou, F. Xing, and Y. Yao. 2011. Abstract next subvolume method: A logical process-based approach for spatial stochastic simulation of chemical reactions. Comput. Biol. Chem. 35, 3 (2011), 193--198. Google ScholarDigital Library
- J. Wang, D. Jagtap, N. B. Abu-Ghazaleh, and D. Ponomarev. 2014. Parallel discrete event simulation for multi-core systems: Analysis and optimization. IEEE Trans. Par. Distr. Syst. 25, 6 (2014), 1574--1584. Google ScholarDigital Library
Index Terms
- Exposing Inter-process Information for Efficient PDES of Spatial Stochastic Systems on Multicores
Recommendations
Exposing Inter-Process Information for Efficient Parallel Discrete Event Simulation of Spatial Stochastic Systems
SIGSIM-PADS '17: Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete SimulationWe present a new efficient approach to the parallelization of discrete event simulators for multicore computers, which is based on exposing and disseminating essential information between processors. We aim specifically at simulation models with a ...
PDES-A: Accelerators for Parallel Discrete Event Simulation Implemented on FPGAs
Special Issue on PADS 2017In this article, we present experiences implementing a general Parallel Discrete Event Simulation (PDES) accelerator on a Field Programmable Gate Array (FPGA). The accelerator can be specialized to any particular simulation model by defining the object ...
Efficient Inter-Process Synchronization for Parallel Discrete Event Simulation on Multicores
SIGSIM PADS '15: Proceedings of the 3rd ACM SIGSIM Conference on Principles of Advanced Discrete SimulationWe present a new technique for controlling optimism in Parallel Discrete Event Simulation on multicores. It is designed to be suitable for simulating models, in which the time intervals between successive events between different processes are highly ...
Comments