Issue No.09 - September (2006 vol.7)
Raphael Y. de Camargo , University of Sao Paulo
Fabio Kon , University of Sao Paulo
Renato Cerqueira , Pontifical Catholic University of Rio de Janeiro
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MDSO.2006.56
This article compares several strategies for storing checkpoint data from parallel applications in an opportunistic grid environment. In terms of computational overhead, storage overhead, and degree of fault tolerance, the authors evaluate the use of replication, parity information, and erasure coding. They use an object-oriented grid middleware solution called InteGrade to implement these strategies and to perform the evaluation experiments.
fault tolerance, distributed storage, data coding, checkpointing, grid computing
Raphael Y. de Camargo, Fabio Kon, Renato Cerqueira, "Strategies for Checkpoint Storage on Opportunistic Grids", IEEE Distributed Systems Online, vol.7, no. 9, pp. 1, September 2006, doi:10.1109/MDSO.2006.56