Flexible Rollback Recovery in Grid Computing

download Flexible Rollback Recovery in Grid Computing

of 22

Transcript of Flexible Rollback Recovery in Grid Computing

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    1/22

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    2/22

    ABSTRACT Grid and cluster architectures is computationally

    intensive parallel applications.

    consisting of computational nodes, mass storage, andinterconnection networks is very complexity.

    Mean Time to Failure (MTTF),decreases the

    complexity.

    Fault tolerance is, thus, a necessity to avoid failure inlarge applications.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    3/22

    PROBLEM IDENTIFICATIONAny process can be check-pointed at any time.

    An alternative approach releases the constraint ofalways check- point table processes.

    This protocol has been implemented within Pro-

    Active.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    4/22

    PROBLEM SOLUTION Fault-tolerance mechanisms called Theft-Induced

    Check pointing and Systematic Event Logging.

    Specifically, the protocols base the state of theexecution on a dataflow graph.

    Efficient recovery in dynamic heterogeneous systemsas well as multithreaded applications.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    5/22

    SYSTEM REQUIREMENTSHardware Requirements: System : Dual Core 2.6GHz.

    Hard Disk : 160 GB.

    Monitor : 15 VGA Colour.

    Mouse : Logitech.

    Ram : 1 GB More.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    6/22

    SYSTEM REQUIREMENTSSoftware Requirements:

    Operating system : - Windows XP Professional.

    Coding Language : - Java,SWING

    Tool Used : - Netbeans 6.9.1

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    7/22

    PROJECT MODULES Network Module

    Logging Module

    Check-pointing Module

    Work Stealing Module

    Fault and Fault Free Module

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    8/22

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    9/22

    LOGGING MODULE Logging can be classified as pessimistic, optimistic, or

    causal.

    Log-based mechanisms in which the onlynondeterministic events in a system are the receptionof messages is usually referred to as message logging.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    10/22

    CHECK-POINTING METHOD

    Checkpointing relies on periodically saving the state of

    the computation to stable storage.

    The consistent global state can be achieved either atthe time of checkpointing or at the time of rollbackrecovery

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    11/22

    WORK-STEALING METHOD The runtime environment and primary mechanism for

    load distribution is a scheduling algorithm calledwork-stealing.

    The principal mechanism for dispatching tasks in thedistributed environment is task stealing.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    12/22

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    13/22

    Server form

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    14/22

    Client A

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    15/22

    Client B

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    16/22

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    17/22

    Client A process

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    18/22

    Client B process

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    19/22

    Client C process

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    20/22

    Conclusion To overcome the problem of applications executing in

    large systems where the MTTF approaches time of theapplication, two fault-tolerant protocols, TIC and SEL,were introduced.

    The experimental results confirmed the theoreticalanalysis and demonstrated the low overhead of bothapproaches.

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    21/22

  • 8/11/2019 Flexible Rollback Recovery in Grid Computing

    22/22