Profiling of Dataflow Programs Using Post Mortem Causation Traces

The natural representation of data streams, parallelism, and composition has made dataflow an attractive programming model for expressing a wide range of stream and media processing applications, and has led MPEG and ISO to base their latest video coding standards on this model. This paper describes and compares methodologies and metrics for the optimization of signal processing algorithms represented as dataflow programs. Our approach is based on the analysis of traces and addresses some of the complexity challenges that arise from the very large data sets that are required for evaluating real-world applications. The methodology and experimental results are demonstrated and evaluated in two at-size case studies, an MPEG-4 SP and an AVC/H.264 video decoders.

[1]  李琳.  The University of California,Berkeley , 2007 .

[2]  Christophe Lucarz,et al.  Optimization of portable parallel signal processing applications by design space exploration of dataflow programs , 2011, 2011 IEEE Workshop on Signal Processing Systems (SiPS).

[3]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[4]  Jörn W. Janneck,et al.  Profiling dataflow programs , 2008, 2008 IEEE International Conference on Multimedia and Expo.

[5]  Barton P. Miller,et al.  Slack: A New Performance Metric for Parallel Programs , 2007 .

[6]  Barton P. Miller,et al.  Critical path analysis for the execution of parallel and distributed programs , 1988, [1988] Proceedings. The 8th International Conference on Distributed.

[7]  Ghislain Roquier,et al.  Scheduling of dynamic dataflow programs with model checking , 2011, 2011 IEEE Workshop on Signal Processing Systems (SiPS).

[8]  Ghislain Roquier,et al.  Hardware and Software Synthesis of Heterogeneous Systems from Dataflow Programs , 2012, J. Electr. Comput. Eng..

[9]  Donna S. Reese,et al.  Near-critical path analysis of program activity graphs , 1994, Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[10]  Christophe Lucarz,et al.  Dataflow Programming for Systems Design Space Exploration for Multicore Platforms , 2011 .

[11]  Ghislain Roquier,et al.  Synthesizing hardware from dataflow programs: An MPEG-4 simple profile decoder case study , 2008, SiPS.

[12]  Jörn W. Janneck.  A machine model for dataflow actors and its applications , 2011, 2011 Conference Record of the Forty Fifth Asilomar Conference on Signals, Systems and Computers (ASILOMAR).

[13]  Jean-François Nezan,et al.  Hardware code generation from dataflow programs , 2010, 2010 Conference on Design and Architectures for Signal and Image Processing (DASIP).

[14]  Barton P. Miller,et al.  Parallel program performance metrics: a comparison and validation , 1992, Proceedings Supercomputing '92.

[15]  Ghislain Roquier,et al.  Scheduling of dynamic dataflow programs based on state space analysis , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[16]  Edward A. Lee,et al.  Taming heterogeneity - the Ptolemy approach , 2003, Proc. IEEE.

[17]  Mickaël Raulet,et al.  An RVC dataflow description of the AVC Constrained Baseline Profile decoder , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[18]  Mickaël Raulet,et al.  The Reconfigurable Video Coding Standard [Standards in a Nutshell] , 2010, IEEE Signal Processing Magazine.