Coordinated Checkpoint versus Message Log for Fault Tolerant MPI

— Large Clusters, high availability clusters and Grid deployments often suffer from network, node or operating system faults and thus require the use of fault tolerant programming models. MPI is one of the most widely adopted programming models for high performance computing. There are several approaches for fault tolerance in an MPI environment. The automatic and transparent ones are based on either coordinated or uncoordinated checkpointing associated with a message log strategy. There are many protocols and optimizations for these approaches and several implementations have been made. However, few results of comparison between them exist. Coordinated checkpoint has the advantage of a very low overhead as long as the execution stays fault free. In contrary, uncoordinated checkpoint must be complemented by a message log protocol which adds a significant penalty for all message transfers even for fault free executions. The drawbacks of coordinated checkpoint are the synchronization ...

• Type: Conference
• Year: 2003
• Where: CLUSTER
• Authors: Aurelien Bouteiller, Pierre Lemarinier, Géraud Krawezik, Franck Cappello
• Keywords: CLUSTER 2003, Cluster Computing, Fault Tolerant Programming, Message Log, Uncoordinated Checkpoint