대한전자공학회논문지SD (Journal of the Institute of Electronics Engineers of Korea SD)

대한전자공학회 (The Institute of Electronics and Information Engineers)
권호 표지

Content-Addressable Memory를 이용한 확장 가능한 범용 병렬 Associative Processor 설계

Design of a scalable general-purpose parallel associative processor using content-addressable memory

  • 박태근 (가톨릭대학교 정보통신전자공학부)
  • Park, Tae-Geun (Department of Information, Communications, and Electronics Engineering, The Catholic University of Korea)
  • 발행 : 2006.02.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

일반 컴퓨터에서 중앙처리장치와 메모리 사이의 병목현상인 "Von Neumann Bottleneck"을 보이는데 본 논문에서는 이러한 문제점을 해소하고 검색위주의 응용분야에서 우수한 성능을 보이는 Content-addressable memory(CAM) 기반의 확장 가능한 범용 Associative Processor(AP) 구조를 제안하였다. 본 연구에서는 Associative computing을 효율적으로 수행할 수 있는 명령어 세트를 제안하였으며 다양하고 대용량 응용분야에도 적용할 수 있도록 구조를 확장 가능하게 설계함으로써 유연한 구조를 갖는다. 12 가지의 명령어가 정의되었으며 프로그램이 효율적으로 수행될 수 있도록 명령어 셋을 구성하고 연속된 명령어를 하나의 명령어로 구현함으로써 처리시간을 단축하였다. 제안된 프로세서는 bit-serial, word-parallel로 동작하며 대용량 병렬 SIMD 구조를 갖는 32 비트 범용 병렬 프로세서로 동작한다. 포괄적인 검증을 위하여 명령어 단위의 검증 뿐 아니라 최대/최소 검색, 이상/이하 검색, 병렬 덧셈 등의 기본적인 병렬 알고리즘을 검증하였으며 알고리즘은 처리 데이터의 개수와는 무관한 상수의 복잡도 O(k)를 갖으며 데이터의 비트 수만큼의 이터레이션을 갖는다.

Von Neumann architecture suffers from the interface between the central processing unit and the memory, which is called 'Von Neumann bottleneck' In this paper, we propose a scalable general-purpose associative processor (AP) based on content-addressable memory (CAM) which solves this problem and is suitable for the search-oriented applications. We propose an efficient instruction set and a structural scalability to extend for larger applications. We define twelve instructions and provide some reduced instructions to speed up which execute two instructions in a single instruction cycle. The proposed AP performs in a bit-serial, word-parallel fashion and can be considered as a 32-bit general-purpose parallel processor with a massively parallel SIMD structure. We design and simulate a maximum/minumum search greater-than/less-than search, and parallel addition to verify the proposed architecture. The algorithms are executed in a constant time O(k) regardless of the number of input data.

키워드

참고문헌

  1. J. Potter, Associative computing: A programming paradigm for massively parallel computers, Plenum publishing, New York, 1992
  2. A. Krikelis, 'Computer vision applications with the associative string processor,' J. of parallel and distributed computing, vol. 13, no. 2, pp. 170-184, 1991 https://doi.org/10.1016/0743-7315(91)90086-O
  3. H. Kitano, Speech-to-speech translation: A massively parallel memory-based approach, Kluwer academic publisher, 1994
  4. C. Weems, 'The image understanding architecture,' Int'l J. computer vision, vol. 2, no. 4, pp.251-282, 1989 https://doi.org/10.1007/BF00158166
  5. R. Storer, et al., 'An associative processing module for a heterogeneous vision architecture,' IEEE Micro, vol. 12, no. 3, pp. 42-55, 1992 https://doi.org/10.1109/40.141602
  6. T. Higuchi, et al., 'The IXM2 parallel associative processor for AI,' IEEE Computer, vol. 27, no. 11, pp. 53-63, 1994 https://doi.org/10.1109/2.330048
  7. A. Louri, et al., 'An optical associative parallel processor for high-speed database processing,' IEEE Computer, vol. 27, no. 11, pp.65-72, 1994 https://doi.org/10.1109/2.330033
  8. K. Grosspietsch and R. Reetz, 'The associative processor system CAPRA: architecture and applications,' IEEE Micro, vol. 12, no. 3, pp. 58-67, 1992 https://doi.org/10.1109/40.180248
  9. D. Tavangarian, 'Flag-oriented parallel associative architectures and applications,' IEEE Computer, vol. 27, no. 11, pp. 41-51, 1994 https://doi.org/10.1109/2.330031
  10. C. Stormon, et al., 'A general-purpose CMOS associative processor IC and system,' IEEE Micro, vol. 12, no. 3, pp. 68-78, 1992 https://doi.org/10.1109/40.180249
  11. K. Tamaru, K. Kobayashi, and H. Onodera, 'Memory based architecture and its implementation scheme named bit-parallel block-parallel functional memory type parallel processor MPMP FMPP,' Comp. and Elect Engineering, vol. 24, pp.17-31, 1998 https://doi.org/10.1016/S0045-7906(97)00039-6
  12. F. Herrmann and C. Sodini, 'A 256-element associative parallel processor,' IEEE J. Solid-State Circuits, pp. 365-370, vol. 30, no. 4, 1995 https://doi.org/10.1109/4.375954