| Document Date: 2014-07-02 04:07:05
Open Document File Size: 555,47 KBShare Result on Facebook
City Honolulu / / Company Yi2 Job1 Group 2 Group / AMD / Xi2 Job3 Group / 4 Group 1 Failure Group / Amdahl / / Country France / United States / / / Facility University of Tennessee Knoxville / / IndustryTerm last / faulttolerance solution / post-petascale high performance computing / bp/gc processors / matrix product / empirical periodic checkpointing solution / exascale systems / classical online list scheduling algorithm / execution protocol / legacy applications / processor groups / numerical search / parallel applications / replication execution protocol / possible solution / runtime systems / software rejuvenation / grid computing / computing / term processor / brute-force algorithm / overall algorithm / real-world systems / volunteer computing / performance computing applications / multicore processor / physical processors / resume computing / / Organization Univ. of Hawai‘i / Ecole Normale Sup´erieure de Lyon / University of Tennessee Knoxville / / Person Vivien Laboratoire / / / Position General / Corresponding author / representative / / ProgrammingLanguage C / V / / Technology ASAP protocol / failed processor / failing processor / multicore processor / overall algorithm / execution protocol / bp/gc processors / 6 Algorithm / second processor / replication execution protocol / using Algorithm / classical online list scheduling algorithm / 7 6.1 Brute-force algorithm The BestPeriod algorithm / DP algorithm / two algorithms / term processor / one processor / resulting algorithm / brute-force algorithm / Dynamic Programming algorithm / participating processor / simulation / 5 Algorithm / 6.2 Dynamic Programming algorithm Algorithm / at least one processor / /
SocialTag |
Using group replication for resilience on exascale systems Marin Bougeret1 , Henri Casanova2 , Yves Robert3,4 , Fr´ed´eric Vivien5,3 and Dounia Zaidouni5,3 1. LIRMM Montpellier, France, 2. Univ.