A Domain-Specific Architecture for Deep Neural Networks.

The article discusses the use of Google's tensor processing units (TPUs) to improve performance per watt of deep neural networks in the company's datacenters, with better energy efficiency than central processing units (CPUs) and graphics processing units (GPUs) in similar technologies.

Detalles Bibliográficos
Publicado en:Communications of the ACM Vol. 61; no. 9; pp. 50 - 60
Autores principales: JOUPPI, NORMAN P., YOUNG, CLIFF, PATIL, NISHANT, PATTERSON, DAVID
Formato: Artículo
Publicado: Association for Computing Machinery Sep2018
Materias:
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=131463997&site=ehost-live
header:
  @attributes:
    shortDbName: hlh
    uiTerm: 131463997
    longDbName: Humanities International Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        00010782
        ACM
      jtl: Communications of the ACM
      issn: 00010782
      maglogo: N
    pubinfo:
      dt: Sep2018
      vid: 61
      iid: 9
      pid: 68
      pub: Association for Computing Machinery
    artinfo:
      ui:
        131463997
        10.1145/3154484
      ppf: 50
      ppct: 10
      formats:
      tig:
        atl: A Domain-Specific Architecture for Deep Neural Networks.
      aug:
        au:
          JOUPPI, NORMAN P.
          YOUNG, CLIFF
          PATIL, NISHANT
          PATTERSON, DAVID
        affil:
          Distinguished Hardware Engineer at Google, Mountain View, CA, USA
          Member of the Brain team at Google, Mountain View, CA, USA
          Tech Lead Manager at Google, Mountain View, CA, USA
          Pardee Professor of Computer Science, Emeritus at the University of California at Berkeley, Berkeley, CA, USA
          Distinguished Engineer at Google, Mountain View, CA, USA
      su:
        Artificial neural networks
        Application-specific integrated circuits
        Server farms (Computer network management)
        Google Inc.
        Energy consumption
      sug:
        subj:
          Artificial neural networks
          Application-specific integrated circuits
          Server farms (Computer network management)
          Google Inc.
          Energy consumption
      ab: The article discusses the use of Google's tensor processing units (TPUs) to improve performance per watt of deep neural networks in the company's datacenters, with better energy efficiency than central processing units (CPUs) and graphics processing units (GPUs) in similar technologies.
      pubtype: Periodical
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: Y
      dt:
        @attributes:
          year: 2018
    holdings:
      @attributes:
        islocal: N