• DocumentCode
    961616
  • Title

    The Associative Linear Array Processor

  • Author

    Finnila, Charles A. ; Love, Hubert H., Jr.

  • Author_Institution
    Strategic Systems Division, Hughes Aircraft Company, Culver City, CA 90230.
  • Issue
    2
  • fYear
    1977
  • Firstpage
    112
  • Lastpage
    125
  • Abstract
    The associative linear array processor (ALAP) is a new approach to making large associative processors practical. Data storage in shift registers, bit-serial arithmetic, LSI word cells comprehensive arithmetic capability within the memory array, and electronic fault isolation are all utilized. The processor is a linear array of word cells, each containing memory and arithmetic logic. All connections to the word cells are by means of common buses except for the multiuse chaining channel. As with most operations within word cells, the current function of the chaining channel in each word is controlled by the combination of common control lines and the states of flag flip-flops in each word. This allows different chaining channel action in different words. Among its many uses the chaining channel can be used to resolve conflicts of multiple associative memory matches, to perform interword arithmetic, or to sort data as they are being input to the associative memory. In the ALAP project effort conducted at Hughes, wafers containing arrays of ALAP cells have been fabricated using LSI technology. A complete hardware/software system using the ALAP has been fabricated and checked out, along with two application programs of nontrivial magnitude.
  • Keywords
    Arithmetic; Associative memory; Circuit faults; Flip-flops; Hardware; Large scale integration; Logic arrays; Logic design; Programmable logic arrays; Shift registers; Associative memory; associative processor system; cellular logic array; data retrieval; electronic fault isolation; fault-tolerant logic; highly parallel arithmetic; large-scale integration (LSI); parallel processing;
  • fLanguage
    English
  • Journal_Title
    Computers, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9340
  • Type

    jour

  • DOI
    10.1109/TC.1977.5009290
  • Filename
    5009290