Design and Implementation of High Performance Content-Addressable Memories.

Abstract

The content-addressable memories (CAMs) have major applications in developing areas of artificial intelligence and image processing. This work involves researching, designing, and implementing a high performance, high density CAM. The design, fabrication, and test of this all-parallel AFIT CAM is discussed. The AFIT CAM stores 2K bits of information and fits in 64 pin dual-in-line package. It was fabricated using a 3 micron CMOS technology with 2-layer metals. The AFIT CAM supports four types of comparisons directly on the chip. These are equal, not-equal, greater-than-or-equal, and less-than. In addition, it has DUMP and Parallel Write operations to aid artificial intelligence and image processing applications. The AFIT CAM has a total of 78,000 devices and achieves a density of 1,073 devices per sq. mm. The memory access time for reading and writing are only 22 ns and 20 ns respectively. The CAM chip computes 800 million magnitude comparisons or 3.2 billion equality comparisons per second. The cif file has been resubmitted to the manufacturer for fabrication due to a short circuit which was caused by over bloated second metal lines. A second version of the AFIT CAM cell has been redesigned and nets a 25% reduction in size. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1985
Accession Number
ADA163995

Entities

People

  • Wook H. Shinn

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies
  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Access Time
  • Artificial Intelligence
  • Circuits
  • Computer Programming
  • Computers
  • Content Addressable Memory
  • Data Storage Systems
  • Detectors
  • Electrical Engineering
  • Engineering
  • High Density
  • Image Processing
  • Inverters
  • Logic Gates
  • Parallel Computing
  • Short Circuits
  • Xor Gates

Readers

  • Aerospace Engineering
  • Integrated Circuit Design and Technology.
  • Parallel and Distributed Computing.

Technology Areas

  • AI & ML