Performance Bounds of Asynchronous Circuits with Mode-Based Conditional Behavior

Asynchronous circuits with conditional behavior often have distinct modes of operation each of which can be modeled as a marked graph with its own performance target. This paper derives performance bounds for such conditional circuits based on the cycle times of successively larger collections of these underlying modes. Our bounds prove the somewhat intuitive result that treating a conditional circuit as unconditional for slack matching guarantees the circuit performance requirement conservatively. We also prove the somewhat counter-intuitive result that the average cycle time of a conditional circuit may be worse than the weighted average of the cycle time of its underlying collection of modes. Finally, the paper outlines the potential application of these bounds to future improvements in slack matching of such conditional circuits.

[1]  Tadao Murata,et al.  Petri nets: Properties, analysis and applications , 1989, Proc. IEEE.

[2]  Steven M. Burns,et al.  Efficient Timing Analysis of a Class of Petri Nets , 1995, CAV.

[3]  David L. Dill,et al.  Efficient algorithms for approximate time separation of events , 2002 .

[4]  Andrew M Lines,et al.  Pipelined Asynchronous Circuits , 1998 .

[5]  Mark R. Greenstreet,et al.  Self-timed meshes are faster than synchronous , 1997, Proceedings Third International Symposium on Advanced Research in Asynchronous Circuits and Systems.

[6]  Peter A. Beerel,et al.  Proteus: An ASIC Flow for GHz Asynchronous Designs , 2011, IEEE Design & Test of Computers.

[7]  Michael Kishinevsky,et al.  Performance Analysis Based on Timing Simulation , 1994, 31st Design Automation Conference.

[8]  Seth Copen Goldstein,et al.  Leveraging protocol knowledge in slack matching , 2006, ICCAD.

[9]  Peter A. Beerel,et al.  Slack matching asynchronous designs , 2006, 12th IEEE International Symposium on Asynchronous Circuits and Systems (ASYNC'06).

[10]  Montek Singh,et al.  Bottleneck Analysis and Alleviation in Pipelined Systems: A Fast Hierarchical Approach , 2009, 2009 15th IEEE Symposium on Asynchronous Circuits and Systems.

[11]  Peter A. Beerel,et al.  Bounding average time separations of events in stochastic timed Petri nets with choice , 1999, Proceedings. Fifth International Symposium on Advanced Research in Asynchronous Circuits and Systems.

[12]  Alexander Taubin,et al.  Heuristic Based throughput Analysis and Optimization of Asynchronous Pipelines , 2009, 2009 15th IEEE Symposium on Asynchronous Circuits and Systems.

[13]  Kenneth Steiglitz,et al.  Bubbles can make self-timed pipelines fast , 1990, J. VLSI Signal Process..

[14]  Peter A. Beerel,et al.  Pipeline optimization for asynchronous circuits: complexity analysis and an efficient optimal algorithm , 2000, IEEE/ACM International Conference on Computer Aided Design. ICCAD - 2000. IEEE/ACM Digest of Technical Papers (Cat. No.00CH37140).

[15]  Jörg Desel,et al.  Partial order behaviour and structure of Petri nets , 2005, Formal Aspects of Computing.

[16]  Vishal Gupta,et al.  Performance estimation and slack matching for pipelined asynchronous architectures with choice , 2008, ICCAD 2008.

[17]  Piyush Prakash,et al.  Slack matching quasi delay-insensitive circuits , 2006, 12th IEEE International Symposium on Asynchronous Circuits and Systems (ASYNC'06).

[18]  Marc Renaudin,et al.  Performance Modeling and Analysis of Asynchronous Linear-Pipeline with Time Variable Delays , 2007, 2007 14th IEEE International Conference on Electronics, Circuits and Systems.

[19]  Peter A. Beerel,et al.  Symbolic techniques for performance analysis of timed systems based on average time separation of events , 1997, Proceedings Third International Symposium on Advanced Research in Asynchronous Circuits and Systems.

[20]  Gaetano Borriello,et al.  An Algorithm for Exact Bounds on the Time Separation of Events in Concurrent Systems , 1995, IEEE Trans. Computers.