Streaming HD H.264 encoder on programmable processors

ABSTRACT

Programmable processors have great advantage over dedicated ASIC design under intense time-to-market pressure. However, real-time encoding of high-definition (HD) H.264 video (up to 1080p) is a challenge to most existing programmable processors. On the other hand, model-based design is widely accepted in developing complex media program. Stream model, an emerging model-based programming method, shows surprising efficiency on many compute-intensive domains especially for media processing. On the basis, this paper proposes a set of streaming techniques for H.264 encoding, and then develops all of the code based on the X264 reference code. Our streaming H.264 encoder is a pure software implementation completely written in high-level language without special hardware/algorithm support. Real execution results show that our encoder achieves significant speedup over the original X264 encoder on various programmable architectures: on X86 Core^TM2 E8200 the speedup is 1.8x, on MIPS 4KEc the speedup is 3.7x, on TMS320 C6416 DSP the speedup is 5.5x, on stream processor STORM-SP16 G220 the speedup is 6.1x. Especially, on STORM processor, the streaming encoder achieves the performance of 30.6 frames per second for a 1080P HD sequence, satisfying the real-time requirement. These indicate that streaming is extremely efficient for this kind of media workload. Our work is also applicable for other media processing applications, and provides architecture insights into dedicated ASIC or FPGA HD H.264 encoders.

REFERENCES

• 1: Thomas Wiegand, "Draft Text of Final Draft International Standard (FDIS) of Joint Video Specification," 7th Meeting of JVT, Pattaya. Mar. 2003.
• 2: W. Thies, et. al., "StreamIt: A language for streaming applications," Proc. Int'l Conference on Compiler Construction, Apr. 2002.
• 3: P. R. Mattson. "A Programming System for the Imagine Media Processor," PhD thesis, Stanford University, 2002.
• 4: J. D. Owens, B. Khailany, et. al, "Comparing reyes and OpenGL on a stream architecture." In HWWS '02: Proceedings of the ACM SIGGRAPH/EU-ROGRAPHICS conference on Graphics hardware, pages 47--56, 2002.
• 5: J. Owens, U. Kapasi, P. Mattson, W. Dally et.al., "Media processing applications on the Imagine stream processor," Proc. IEEE Int'l Conference on Computer Design, Sep. 2002.
• 6: Hyeyoung Hwang Taewook Oh, Hyunuk Jung, Soonhoi Ha, "Conversion of Reference C Code to Dataflow Model:H.264 Encoder Case Study," ASP-DAC, 2006.
• 7: J.Gummaraju and M. Rosenblum. "Stream Programming on General-Purpose Processors," Proc. 38th annual ACM/IEEE Int'l Symposium on Microarchitecture, Barcelona, Spain, November 2005.
• 8: F. Labonte, et.al, "The Stream Virtual Machine", Proc. 2004 Int'l Conference on Parallel Architectures and Compilation Techniques, France, September 2004.
• 9: Reference software X264-060805. http://www.videolan.org/developers/x264.html
• 10: Browne, S., Deane, C., Ho, G., Mucci, P. "PAPI: A Portable Interface to Hardware Performance Counters," Proc. Department of Defense HPCMP Users Group Conference, June 1999.
• 11: Tiago A. da Fonseca, Yuxin Liu, and Ricardo L. de Queiroz, "Open-Loop Prediction in H.264/AVC for High Definition Sequences," Proc. SBrT 2007, Blumenau,Brazil March 2007.
• 12: Ishfaq Ahmad , Yong He , Ming L. Liou, "Video Compression with Parallel Processing," Parallel Computing, v.28 n.7-8, p.1039--1078, August 2002
• 13: Mei Wen, et al., "On-chip Memory System Optimization Design for the FT64 Scientific Stream Accelerator", IEEE Micro, July/August 2008
• 14: I. Buck, T. Foley, D. Horn, et.al., "Brook for GPUs: Stream computing on graphics hardware," Proc. SIGGRAPH, 2004.
• 15: Stream Processors Inc. "SPI Software Documentation," http://www.streamprocessors.com, 2008

INDEX TERMS

Primary Classification:
  D. Software
  D.1 PROGRAMMING TECHNIQUES
      D.1.3 Concurrent Programming
          Subjects: Parallel programming

General Terms:
Algorithms, Performance

Keywords:
1080P HD, H.264 encoder, programmable, real-time, stream