• DocumentCode
    1614736
  • Title

    A horizontal optimization algorithm for data path/control synthesis

  • Author

    Peng, Zebo

  • Author_Institution
    Dept. of Comput. & Inf. Sci., Linkoping Univ., Sweden
  • fYear
    1988
  • Firstpage
    239
  • Abstract
    Describes an approach to the synthesis of VLSI systems from their high-level behavioral descriptions into structural implementations. This approach integrates operation scheduling, data-path allocation, and control allocation of a traditional synthesis process. Therefore, the advantages from compacting the data path, for example, can be compared immediately with the possible expansion of the control logic to justify the compaction. The horizontal algorithm is designed to make decisions on what should be done in each synthesis step. When making such design decisions, it makes tradeoffs between the data path and control, resource sharing, and operation scheduling, as well as cost and performance. The experimental results show that under the same conditions, a horizontal approach performs better than a vertical approach with respect to overall system performance and estimated layout area. The proposed horizontal optimization algorithm was designed and implemented in computer-aided design and manufacturing
  • Keywords
    CAD/CAM; VLSI; circuit CAD; CAM; compacting; computer-aided design; control logic; data-path allocation; high-level behavioral descriptions; horizontal optimization algorithm; layout area; operation scheduling; overall system performance; resource sharing; structural implementations; tradeoffs; Algorithm design and analysis; Compaction; Control system synthesis; Costs; Job shop scheduling; Logic; Processor scheduling; Resource management; System performance; Very large scale integration;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems, 1988., IEEE International Symposium on
  • Conference_Location
    Espoo
  • Type

    conf

  • DOI
    10.1109/ISCAS.1988.14911
  • Filename
    14911