A survey of rollback-recovery protocols in message-passing systems
ACM Computing Surveys (CSUR)
Message Logging in Mobile Computing
FTCS '99 Proceedings of the Twenty-Ninth Annual International Symposium on Fault-Tolerant Computing
Publishing: a reliable broadcast communication mechanism
SOSP '83 Proceedings of the ninth ACM symposium on Operating systems principles
Sender-based message logging for reducing rollback propagation
SPDP '95 Proceedings of the 7th IEEE Symposium on Parallel and Distributeed Processing
Future Generation Computer Systems - Special issue: Advanced services for clusters and internet computing
MPICH-V2: a Fault Tolerant MPI for Volatile Nodes based on Pessimistic Sender Based Message Logging
Proceedings of the 2003 ACM/IEEE conference on Supercomputing
| Hi-index | 0.00 |
Among message logging approaches, volatile logging by sender processes considerably alleviates the normal operation overhead of synchronous logging on stable storage. But, this approach forces each process to maintain log information of its sent messages in its limited volatile memory. This paper presents a novel log management algorithm to effectively eliminate the volatile log information at sender processes on demand without the violation of the system consistency. First, the algorithm gets rid of needless logged messages from the corresponding senders' volatile memories only by piggybacking a vector on their sent messages. This advantageous feature results in no additional message and forced checkpoint. If the more empty buffer space for the volatile logging is needed after the first procedure executed, the next procedure of this proposed algorithm is performed to address this limitation. This procedure uses a vector for saving the size of the log information required to recover every other process and enables the information to be efficiently removed while satisfying the consistency condition.