2006 · 89 citations · 18 references
ComputingComputer ScienceHigh-performance ComputingFault ToleranceFault-tolerant ComputingParallel ComputingFault Tolerance DesignFault Tolerance In Parallel SystemsMessage LoggingCheckpointingCoordinated CheckpointingFailure ProbabilityHigh-performance Computing SystemsLoggingParallel Computing ArchitecturesParallel Computing PerformanceScalability
A long-term trend in high-performance computing is the increasing number of nodes in parallel computing platforms, which entails a higher failure probability. Fault tolerant programming environments should be used to guarantee the safe execution of critical applications. Research in fault tolerant MPI has led to the development of several fault tolerant MPI environments. Different approaches are being proposed using a variety of fault tolerant message passing protocols based on coordinated checkpointing or message logging. The most popular approach is with coordinated checkpointing. In the literature, two different concepts of coordinated checkpointing have been proposed: blocking and nonblocking. However they have never been compared quantitatively and their respective scalability remains unknown. The contribution of this paper is to provide the first comparison between these two approaches and a study of their scalability. We have implemented the two approaches within the MPICH environments and evaluate their performance using the NAS parallel benchmarks.
18
K. Mani Chandy, Leslie Lamport · ACM Transactions on Computer Systems · 1985 · 2.4K citations · Full text
System structure for software fault tolerance
Brian Randell · ACM SIGPLAN Notices · 1975 · 1.5K citations · Full text
System structure for software fault tolerance
Brian Randell · IEEE Transactions on Software Engineering · 1975 · 1.1K citations
Optimistic recovery in distributed systems
Rob Strom, Shaula Yemini · ACM Transactions on Computer Systems · 1985 · 722 citations · Full text