Three-dimensional memory vectorization for high bandwidth media memory systems

ABSTRACT

Vector processors have good performance, cost and adaptability when targeting multimedia applications. However, for a significant number of media programs, conventional memory configurations fail to deliver enough memory references per cycle to feed the SIMD functional units. This paper addresses the problem of the memory bandwidth.We propose a novel mechanism suitable for 2-dimensional vector architectures and targeted at providing high effective bandwidth for SIMD memory instructions. The basis of this mechanism is the extension of the scope of vectorization at the memory level, so that 3-dimensional memory patterns can be fetched into a second-level register file.By fetching long blocks of data and by reusing 2-dimensional memory streams at this second-level register file, we obtain a significant increase in the effective memory bandwidth. As side benefits, the new 3-dimensional load instructions provide a high robustness to memory latency and a significant reduction of the cache activity, thus reducing power and energy requirements. At the investment of a 50% more area than a regular SIMD register file, we have measured and average speed-up of 13% and the potential for power savings in the L2 cache of a 30%.

REFERENCES

• 1: Keith Diefendorff , Pradeep K. Dubey, How Multimedia Workloads Will Change Processor Design, Computer, v.30 n.9, p.43-45, September 1997 [doi>10.1109/2.612247]
• 2: Rob Koenen, MPEG-4: multimedia for our time, IEEE Spectrum, v.36 n.2, p.26-33, Feb. 1999 [doi>10.1109/6.744873]
• 3: Alex Peleg , Uri Weiser, MMX Technology Extension to the Intel Architecture, IEEE Micro, v.16 n.4, p.42-50, August 1996 [doi>10.1109/40.526924]
• 4: Pentium iii processor: Developer's manual. Technical Report http://developer.intel.com/design/PentiumlIl, INTEL, 1999.
• 5: Marc Tremblay , J. Michael O'Connor , Venkatesh Narayanan , Liang He, VIS Speeds New Media Processing, IEEE Micro, v.16 n.4, p.10-20, August 1996 [doi>10.1109/40.526921]
• 6: 3dnow! technology manual. Technical Report http://www.amd.com, Advanced Micro Devices, Inc., 1999.
• 7: Mips extension for digital media with 3d. Technical Report http://www.mips.com, MIPS technologies, Inc., 1997.
• 8: Keith Diefendorff , Pradeep K. Dubey , Ron Hochsprung , Hunter Scales, AltiVec Extension to PowerPC Accelerates Media Processing, IEEE Micro, v.20 n.2, p.85-95, March 2000 [doi>10.1109/40.848475]
• 9: Parthasarathy Ranganathan , Sarita Adve , Norman P. Jouppi, Performance of image and video processing with general-purpose processors and media ISA extensions, Proceedings of the 26th annual international symposium on Computer architecture, p.124-135, May 01-04, 1999, Atlanta, Georgia, United States
• 10: Jesus Corbal , Mateo Valero , Roger Espasa, Exploiting a new level of DLP in multimedia applications, Proceedings of the 32nd annual ACM/IEEE international symposium on Microarchitecture, p.72-79, November 16-18, 1999, Haifa, Israel
• 11: Ben H. H. Juurlink , Stamatis Vassiliadis , Dmitri Tcheressiz , Harry A. G. Wijshoff, Implementation and Evaluation of the Complex Streamed Instruction Set, Proceedings of the 2001 International Conference on Parallel Architectures and Compilation Techniques, p.73-82, September 08-12, 2001
• 12: Ahmed El-Mahdy , Ian Watson, A Two Dimensional Vector Architecture for Multimedia, Proceedings of the 7th International Euro-Par Conference Manchester on Parallel Processing, p.687-696, August 28-31, 2001
• 13: Nathan T. Slingerland , Alan Jay Smith, Cache performance for multimedia applications, Proceedings of the 15th international conference on Supercomputing, p.204-217, June 2001, Sorrento, Italy [doi>10.1145/377792.377833]
• 14: Toni Juan , Juan J. Navarro , Olivier Temam, Data caches for superscalar processors, Proceedings of the 11th international conference on Supercomputing, p.60-67, July 07-11, 1997, Vienna, Austria [doi>10.1145/263580.263595]
• 15: R. E. Kessler, The Alpha 21264 Microprocessor, IEEE Micro, v.19 n.2, p.24-36, March 1999 [doi>10.1109/40.755465]
• 16: Francisca Quintana , Jesus Corbal , Roger Espasa , Mateo Valero, Adding a vector unit to a superscalar processor, Proceedings of the 13th international conference on Supercomputing, p.1-10, June 20-25, 1999, Rhodes, Greece [doi>10.1145/305138.305148]
• 17: Chunho Lee , Miodrag Potkonjak , William H. Mangione-Smith, MediaBench: a tool for evaluating and synthesizing multimedia and communicatons systems, Proceedings of the 30th annual ACM/IEEE international symposium on Microarchitecture, p.330-335, December 01-03, 1997, Research Triangle Park, North Carolina, United States
• 18: Jesus Corbal, Roger Espasa, and Mateo Valero. Morn: Instruction set architecture. Technical report, Universitat Politècnica de Catalunya, 1999.
• 19: Amitabh Srivastava , Alan Eustace, ATOM: a system for building customized program analysis tools, Proceedings of the ACM SIGPLAN 1994 conference on Programming language design and implementation, p.196-205, June 20-24, 1994, Orlando, Florida, United States
• 20: S. Rixner, W.J. Dally, B. Khailany, P. Mattson, U. Kapasi, and J.D. Owens. Register organization for media processing. High Performance Computer Architecture, HPCA-5, pages 375--386, 2000.
• 21: Peter Bannon. Alpha 21364: A Scalable Single-chip SMP. Technical Report http://www.digital.com/alphaoem/microprocessorforum.htm, Compaq Computer Corporation, 1998.
• 22: Jesus Corbal , Roger Espasa , Mateo Valero, DLP +TLP Processors for the Next Generation of Media Workloads, Proceedings of the 7th International Symposium on High-Performance Computer Architecture, p.219, January 20-24, 2001
• 23: Christoforos Kozyrakis, A Media-Enhanced Vector Architecture for Embedded Memory Systems, University of California at Berkeley, Berkeley, CA, 1999
• 24: William J. Dally. Tomorrow's computing engines (keynote speech). Feb 1998.
• 25: Atsushi Kunimatsu , Nobuhiro Ide , Toshinori Sato , Yukio Endo , Hiroaki Murakami , Takayuki Kamei , Masashi Hirano , Fujio Ishihara , Haruyuki Tago , Masaaki Oka , Akio Ohba , Teiji Yutaka , Toyoshi Okada , Masakazu Suzuoki, Vector Unit Architecture for Emotion Synthesis, IEEE Micro, v.20 n.2, p.40-47, March 2000 [doi>10.1109/40.848471]
• 26: Cost-Effective Hardware Acceleration of Multimedia Applications, Proceedings of the International Conference on Computer Design: VLSI in Computers & Processors, p.415, September 23-26, 2001
• 27: Francky Catthoor , Eddy de Greef , Sven Suytack, Custom Memory Management Methodology: Exploration of Memory Organisation for Embedded Multimedia System Design, Kluwer Academic Publishers, Norwell, MA, 1998
• 28: R. Schaffer, F. Catthoor, and R. Merker. Combining background memory management and regular array co-partitioning illustrated on a full motion estimation kernel. special issue on Advanced Regular Array Design (T. Plaks, ed.) in J. of Parallel Algorithms and Applications, Vol. 15, No. 3-4:pp. 201--228, December 2000.
• 29: Lixin Zhang , John B. Carter , Wilson C. Hsieh , Sally A. McKee, Memory System Support for Image Processing, Proceedings of the 1999 International Conference on Parallel Architectures and Compilation Techniques, p.98, October 12-16, 1999
• 30: Krste Asanovic , John Wawrzynek, Vector microprocessors, 1998
• 31: Akihiro Iwaya and Tadashi Watanabe. The parallel processing feature of the NEC SX-3 supercomputer system. Intl. Journal of High Speed Computing, 3(3&4):187--197, 1991.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.1 PROCESSOR ARCHITECTURES
      C.1.2 Multiple Data Stream Architectures (Multiprocessors)
          Subjects: Array and vector processors

Additional Classification:
  B. Hardware
  B.3 MEMORY STRUCTURES
      B.3.2 Design Styles
          Subjects: Cache memories

General Terms:
Performance