• DocumentCode
    1970091
  • Title

    First failure data capture in embedded system

  • Author

    Sun, Hai Tao

  • Author_Institution
    IBM Corp., Rochester
  • fYear
    2007
  • fDate
    17-20 May 2007
  • Firstpage
    183
  • Lastpage
    187
  • Abstract
    First failure data capture (FFDC) is a technology that is used to capture diagnostic information when errors occur. This information reduces the need to reproduce errors. FFDC involves all firmware components. The processes and techniques of FFDC are crucial to the success of an embedded system. With a good FFDC implementation, software engineers can collect diagnostic data quickly and fix defects without impacting customer´s operation. One of the challengers with FFDC is collecting the useful data. There are cases where the volume of data collected is huge but only covers a short amount of time. In other cases, the collected data cover a long period time, but only gather a few variables or items. A new FFDC technique has been developed to capture useful data. This new FFDC technique captures very detail data for a short period of time and captures much less detail data for a long period of time. This paper first introduces three FFDC techniques. The three techniques are general dump, event trace and error log. Second, it analyzes the problems of current FFDC techniques. Third, it discusses a new FFDC technique. This new FFDC technique is low cost, efficient and good for data collection.
  • Keywords
    embedded systems; error handling; program debugging; program diagnostics; diagnostic information; embedded system; error log; event trace; firmware components; first failure data capture; general dump; Computer errors; Costs; Data engineering; Embedded software; Embedded system; Engineering management; Hardware; Microprogramming; Road transportation; Sun; First failure data capture; embedded system; multiple trace buffer;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electro/Information Technology, 2007 IEEE International Conference on
  • Conference_Location
    Chicago, IL
  • Print_ISBN
    978-1-4244-0941-9
  • Electronic_ISBN
    978-1-4244-0941-9
  • Type

    conf

  • DOI
    10.1109/EIT.2007.4374434
  • Filename
    4374434