MPICH-V2: a Fault Tolerant MPI for Volatile Nodes based on Pessimistic Sender Based Message Logging.
Aurélien BouteillerFranck CappelloThomas HéraultGéraud KrawezikPierre LemarinierFrédéric MagniettePublished in: SC (2003)
Keyphrases
- fault tolerant
- fault tolerance
- distributed systems
- interconnection networks
- message passing
- load balancing
- parallel algorithm
- high performance computing
- safety critical
- parallel implementation
- shared memory
- message exchange
- communication channels
- shortest path
- parallel computing
- high availability
- expert systems
- state machine
- short messages
- data structure
- sensor nodes
- software systems
- digital libraries