Hi! I want the detailed advantages and disadvantages of Flexible Rollback Recovery in Dynamic Heterogeneous Grid Computing(DotNet Project)

Flexible Rollback Recovery in
Dynamic Heterogeneous Grid Computing
Large applications executing on Grid or cluster architectures
consisting of hundreds or thousands of computational
nodes create problems with respect to reliability.node failures and the need for dynamic
configuration over extensive run-time maybe the sources of the problem. Theft Induced Checkpointing
and Systematic Event Logging methods for fault-tolerance are described here.

INTRODUCTION
GRID and cluster architectures have gained popularity for
computationally intensive parallel applications.The high failure probabilities are due to the fact
that, in the absence of fault-tolerance mechanisms, the failure of
a single node will cause the entire execution to fail.characteristics of a heterogeneous and dynamic environment has to be considered by the fault tolerance mechanism. Besides heterogeneity one has to address the dynamic nature
of the Grid. configuration changes
may be induced by the application as the result of changes of runtime
observable Quality of Service (QoS) parameters.flexible rollback recovery mechanisms that impose no
artificial restrictions on the execution is described here. They consider node and cluster failures
as well as operation-induced unavailability of resources and dynamic topology reconfiguration and do not depend on the
pre-failure configuration of the heterogeneous network.

BACKGROUND
there are two disadvantages associated with redundancy:
1)Only a fixed number of faults can be tolerated depending on
the type of fault.
2)The necessary degree of redundancy may introduce unacceptable
cost associated with the redundant parallel computations.
In information redundancy redundant information
is added that can be used during recovery to reconstruct
the original data or computation.

Logging-based Approaches:
It is based on the fact that the execution of a process can
be modeled as a sequence of state intervals and The execution during
a state interval is deterministic. It can be classified as pessimistic, optimistic or
causal.

Checkpointing-based Approaches:
checkpointing relies on periodically
saving the state of the computation to stable storage. the computation is restarted from one of the previously
saved states if a fault occurs. In this way checkpointing based
methods differ in the way processes are coordinated and in
the derivation of a consistent global state.

EXECUTION MODEL
The general execution model of large Grid applications can be
considered as having two levels:
Level 0:
only
creates the abstraction of the execution state of the application.Here, the program to be executed is viewed as an
abstraction that represents the state symbolizing the future of
an execution.The input to the virtual machine is the sequential input program added with instructions for the
run-time system.

Level 1:
The abstraction created by the level 0 is then used here to actually schedule and
execute the workload.It schedules tasks using
the primitives Task Export, Task Import and Task Execution.This Level implements the dispatcher, whose decisions will be executed at Level 0.it does not depend on the number
of resources like processors.

Work-stealing
The principle of this algorithm is when a process
becomes idle it tries to steal work from another process called
victim. The initiating process is called thief.The run-time environment and primary mechanism for load
distribution is based on this scheduling algorithm.

THEFT INDUCED CHECKPOINTING
Definition of a checkpoint:checkpoints
are with respect to a process, and consist of a copy of its local
Gi, representing the processâ„¢ stack.

Checkpoint protocol definition:
checkpoints creation can be initiated by either workstealing
or at specific checkpointing periods.Checkpoints resulting from work-stealing are
called forced checkpoints. local checkpoints, are the checkpoints which are stored periodically.
after expiration of pre-defined periods.

full report download:
[attachment=1169]

Thank you..
It is very useful to me..............:-)

Hey,
We feel so nice to hear that it was useful to you. Keep visiting our site and please tell your friends too!

hi, i want detail information abt flexibility rollback recovery in dynamic heterogeneous grid computing and paper prentation and slides of that please send

Large applications executing on Grid or cluster architectures consisting of hundreds or thousands of computational nodes create problems with respect to reliability. The source of the problems are node failures and the need for dynamic configuration over extensive runtime. This paper presents two fault-tolerance mechanisms called Theft-Induced Checkpointing and Systematic Event Logging. These are transparent protocols capable of overcoming problems associated with both benign faults, i.e., crash faults, and node or subnet volatility. Specifically, the protocols base the state of the execution on a dataflow graph, allowing for efficient recovery in dynamic heterogeneous systems as well as multithreaded applications. By allowing recovery even under different numbers of processors, the approaches are especially suitable for applications with a need for adaptive or reactionary configuration control. The low-cost protocols offer the capability of controlling or bounding the overhead. A formal cost model is presented, followed by an experimental evaluation. It is shown that the overhead of the protocol is very small, and the maximum work lost by a crashed process is small and bounded.

please read http://studentbank.in/report-Flexible-Ro...ull-report for more about Flexible Rollback Recovery in Dynamic Heterogeneous Grid Computing

please upload or send ppt topic in flexible rollback recovery on dynamic heterogrneous grid computing

ppt is not available as of now. We will upload it soon as we get it and let you know

for getting more info please visit this page link :-

http://cpe.virginia.edu/pdf/Flexible.pdf