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. The source of the problems are node failures and the need for dynamic configuration over extensive run-time. This paper presents two f...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on dependable and secure computing 2009-01, Vol.6 (1), p.32-44
Main Authors: Jafar, S., Krings, A., Gautier, T.
Format: Article
Language:English
Subjects:
Adaptive control
Checkpointing
Computation
Computer architecture
Computer crashes
Computer Science
Costs
Dataflow
Digital Object Identifier
Distributed architectures
Distributed processing
Dynamical systems
Dynamics
Failure
Fault tolerance
Fault tolerant systems
Faults
Grid computing
Information control
Infrastructure
Operating systems
Other
Programmable control
Protocols
Recovery
Run time (computers)
Runtime
Studies
Volatility
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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 run-time. 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 multi-threaded applications. By allowing recovery even under different numbers of processors, the approaches are especially suitable for applications with 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.
ISSN:1545-5971
1941-0018
DOI:10.1109/TDSC.2008.17