Partial order method for timed simulation of system-level MPSoC designs

ABSTRACT

Current discrete event simulator requires heavy simulation overhead to switch between different components to simulate them in strictly chronological order. Therefore, timed simulation is significantly slower than un-timed simulation. By simply adding delays in the components and communication channels, our timed MPEG-2 decoder simulates more than 14 times slower than an un-timed simulation. In this paper, we propose a partial order method to speed up timed simulation by relaxing the order that the components are simulated. With partial order method, a component is not required to schedule a channel access if both behavioral and timing results of the access are known. The simulation switches less frequently hence the simulation overhead reduces. We show that partial order method can be used in complex system-level simulation such as MPSoC implementations of the MPEG-2 decoder. In our experiments, partial order method provides more than 10 times speedups over regular discrete event simulation for timed simulation.

REFERENCES

• 1: Advanced RISC Machine. http://www.arm.com.
• 2: Felice Balarin , Yosinori Watanabe , Harry Hsieh , Luciano Lavagno , Claudio Passerone , Alberto Sangiovanni-Vincentelli, Metropolis: An Integrated Electronic System Design Environment, Computer, v.36 n.4, p.45-52, April 2003 [doi>10.1109/MC.2003.1193228]
• 3: Luca Benini , Davide Bertozzi , Alessandro Bogliolo , Francesco Menichelli , Mauro Olivieri, MPARM: Exploring the Multi-Processor SoC Design Space with SystemC, Journal of VLSI Signal Processing Systems, v.41 n.2, p.169-182, September 2005 [doi>10.1007/s11265-005-6648-1]
• 4: K. M. Chandy , J. Misra, Asynchronous distributed simulation via a sequence of parallel computations, Communications of the ACM, v.24 n.4, p.198-206, April 1981 [doi>10.1145/358598.358613]
• 5: E. Cheung, H. Hsieh, and F. Balarin. Framework for fast and accurate performance simulation of multiprocessor systems. In High Level Design Validation and Test Workshop, 2007. HLVDT 2007. IEEE International, pages 21--28, Irvine, CA, Nov. 2007.
• 6: Jérôme Cornet , Florence Maraninchi , Laurent Maillet-Contoz, A method for the efficient development of timed and untimed transaction-level models of systems-on-chip, Proceedings of the conference on Design, automation and test in Europe, March 10-14, 2008, Munich, Germany [doi>10.1145/1403375.1403381]
• 7: Poletti Francesco , Paul Marchal , David Atienza , Luca Benini , Francky Catthoor , Jose M. Mendias, An integrated hardware/software approach for run-time scratchpad management, Proceedings of the 41st annual conference on Design automation, June 07-11, 2004, San Diego, CA, USA [doi>10.1145/996566.996634]
• 8: Daniel D. Gajski , Rainer Domer , Junyu Peng , Andreas Gerstlauer, System Design: A Practical Guide with Specc, Kluwer Academic Publishers, Norwell, MA, 2001
• 9: Michael I. Gordon , William Thies , Saman Amarasinghe, Exploiting coarse-grained task, data, and pipeline parallelism in stream programs, Proceedings of the 12th international conference on Architectural support for programming languages and operating systems, October 21-25, 2006, San Jose, California, USA
• 10: John L. Hennessy , David A. Patterson, Computer architecture (2nd ed.): a quantitative approach, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1996
• 11: Suneel Jain, Circular scheduling: a new technique to perform software pipelining, ACM SIGPLAN Notices, v.26 n.6, p.219-228, June 1991
• 12: David R. Jefferson, Virtual time, ACM Transactions on Programming Languages and Systems (TOPLAS), v.7 n.3, p.404-425, July 1985 [doi>10.1145/3916.3988]
• 13: G. Kahn. The semantics of a simple language for parallel programming. In J. L. Rosenfeld, editor, Information processing, pages 471--475, Stockholm, Sweden, Aug 1974. North Holland, Amsterdam.
• 14: Microprocessor without Interlocked Pipeline Stages. http://www.mips.com.
• 15: Open SystemC Initiative. http://www.systemc.com.
• 16: Daniel Gracia Pérez , Gilles Mouchard , Olivier Temam, A New Optimized Implemention of the SystemC Engine Using Acyclic Scheduling, Proceedings of the conference on Design, automation and test in Europe, p.10552, February 16-20, 2004
• 17: Gunar Schirner , Rainer Dömer, Introducing preemptive scheduling in abstract RTOS models using result oriented modeling, Proceedings of the conference on Design, automation and test in Europe, March 10-14, 2008, Munich, Germany [doi>10.1145/1403375.1403408]
• 18: Tensilcia. http://www.tensilica.com.