Trace-driven studies of VLIW video signal processors

This paper uses extensive traces to explore the parallel architectures of highly programmable Video Signal Processors (VSPs). First we briefly compare some existing architectures. Based on the parallel property of video applications, a general Very Long Instruction Word (VLIW) paradigm is proposed. Then we use a new technique, trace-driven architectural exploration, to evaluate architectures with different parameters such as number of registers, number and type of functional units, and number of pipeline stages, et al. Focusing on video applications, we have analyzed the traces of H.263, MPEG-2 and MPEG-4 video codecs. The results show us which architectural tradeoffs enhance the overall performance in the application domain and how we should balance processor resources among registers, functional units and memory units.

[1]  Peter Pirsch,et al.  Array architectures for block matching algorithms , 1989 .

[2]  Wayne Wolf,et al.  Parallelism analysis of the memory system in single-chip VLIW video signal processors , 1998, Electronic Imaging.

[3]  Ruby B. Lee.  Subword parallelism with MAX-2 , 1996, IEEE Micro.

[4]  Joseph A. Fisher,et al.  Trace Scheduling: A Technique for Global Microcode Compaction , 1981, IEEE Transactions on Computers.

[5]  Andrew Wolfe,et al.  Design methodology for programmable video signal processors , 1997, Electronic Imaging.

[6]  Sumedh W. Sathaye,et al.  Instruction fetch mechanisms for VLIW architectures with compressed encodings , 1996, Proceedings of the 29th Annual IEEE/ACM International Symposium on Microarchitecture. MICRO 29.

[7]  Wayne H. Wolf,et al.  Code compression for embedded systems , 1998, Proceedings 1998 Design and Automation Conference. 35th DAC. (Cat. No.98CH36175).

[8]  Takao Nishitani,et al.  A microprogrammable real-time video signal processor (VSP) for motion compensation , 1988 .

[9]  Craig Hansen.  Architecture of a broadband mediaprocessor , 1996, COMPCON '96. Technologies for the Information Superhighway Digest of Papers.

[10]  Stephen Purcell.  Mpact 2 media processor: balanced 2X performance , 1997, Electronic Imaging.

[11]  Andrew Wolfe,et al.  Datapath design for a VLIW video signal processor , 1997, Proceedings Third International Symposium on High-Performance Computer Architecture.

[12]  Ming-Ting Sun,et al.  An all-ASIC implementation of a low bit-rate video codec , 1992, IEEE Trans. Circuits Syst. Video Technol..