• DocumentCode
    2353953
  • Title

    Deadline-aware scheduling for Software Transactional Memory

  • Author

    Maldonado, Walther ; Marlier, Patrick ; Felber, Pascal ; Lawall, Julia ; Muller, Giller ; Rivière, Etienne

  • Author_Institution
    Univ. of Neuchatel, Neuchatel, Switzerland
  • fYear
    2011
  • fDate
    27-30 June 2011
  • Firstpage
    257
  • Lastpage
    268
  • Abstract
    Software Transactional Memory (STM) is an optimistic concurrency control mechanism that simplifies the development of parallel programs. Still, the interest of STM has not yet been demonstrated for reactive applications that require bounded response time for some of their operations. We propose to support such applications by allowing the developer to annotate some transaction blocks with deadlines. Based on previous execution statistics, we adjust the transaction execution strategy by decreasing the level of optimism as the deadlines near through two modes of conservative execution, without overly limiting the progress of concurrent transactions. Our implementation comprises a STM extension for gathering statistics and implementing the execution mode strategies. We have also extended the Linux scheduler to disable preemption or migration of threads that are executing transactions with deadlines. Our experimental evaluation shows that our approach significantly improves the chance of a transaction meeting its deadline when its progress is hampered by conflicts.
  • Keywords
    Linux; concurrency control; multi-threading; scheduling; shared memory systems; transaction processing; Deadline-Aware Scheduling; Linux scheduler; Software Transactional Memory; concurrency control; conservative execution; parallel programs; thread migration; transaction execution strategy; Games; Instruction sets; Linux; Optimized production technology; Rendering (computer graphics); Reservoirs; Runtime; Contention Management; Scheduling; Transactional Memory;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Dependable Systems & Networks (DSN), 2011 IEEE/IFIP 41st International Conference on
  • Conference_Location
    Hong Kong
  • ISSN
    1530-0889
  • Print_ISBN
    978-1-4244-9232-9
  • Electronic_ISBN
    1530-0889
  • Type

    conf

  • DOI
    10.1109/DSN.2011.5958224
  • Filename
    5958224