… correct result, and HPC applications that must guarantee time … a good but inefficient
fault-tolerant solution. We evaluate the … up until where we identify the design’s most sensitive and …

… This work addresses the challenge of designing a reliable replication protocol that provides
both … Failure model We consider a partially synchronous system [34] where processes are …

… The fault tolerant algorithm Theproposedmodelhandlesfaultsthatmayoccurinallavailablevirtual…
and parallel and distributed algorithms design for sensor networks, clusters and grids. …

… We design and implement a resilience pragma to support … Our fault model considers soft
errors/SDCs that materialize in … it the need for task-based replication. Unlike our work, they do …

… as design considerations in current and future HPC systems. … of the art on relevant fault
models for HPC systems (Section 2… to be “scalable, efficient, fault tolerant and easy-to-manage” […

… Designing and modelling selective replication for fault-tolerant HPC applications. In: 17th
IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID), pp. 452…

… in the development of reliable, fault-tolerant GPU architectures. Our … HPC applications.
The frequency of these faults in clustered … We propose and evaluate selective fault tolerance …

… and their extensions and discuss the design choices employed by the … sphere of replication,
environment, hardware model, or … [71] observe that the proposed fault-tolerant CMPs suffer …

… The most basic form of resilience is replication, whereby … This section contains illustrative
applications of the fault-tolerant … design patterns across the layers of HPC systems and …

… Fault tolerant designs are provided to protect the remaining portion of the die covering CPU
… In strict model of replication, Reunion incurs an average performance penalty of 5% and 2% …