Research and applications of clusters of workstations are growing rapidly. One of the major area is fault tolerance. This report describes two issues concerned: correctness and performance. After a number of techniques to improve performance are described, new research directions, diskless checkpointing and Java checkpointing, are introduced.

Checkpointing and rollback recovery are well-known techniques for handling failures in distributed database systems. In this paper, we establish the necessary and sufficient conditions for the checkpoints on a set of data items to be part of a transaction-consistent global checkpoint of the distributed database. This can throw light on designing efficient, non-intrusive checkpointing techniques...

We describe a competent base checkpointing code generation tool. Scientific applications that require long computation runtimes experience the risk of hardware or software failures. Failure during a long computation requires that the application restart the calculations from the beginning. Checkpointing important data at set intervals and using that data during a restart of the application mini...

Scientific workflows are dataand compute-intensive; thus, they may run for days or even weeks on parallel and distributed infrastructures such as grids, supercomputers, and clouds. In these high-performance computing infrastructures, the number of failures that can arise during scientific-workflow enactment can be high, so the use of fault-tolerance techniques is unavoidable. The most-frequentl...

Checkpointing is one of the commonly used techniques to provide fault-tolerance in distributed systems so that the system can operate even if one or more components have failed. However, mobile computing systems are constrained by low bandwidth, mobility, lack of stable storage, frequent disconnections and limited battery life. Hence, checkpointing protocols having lesser number of synchronizat...

Selecting optimal intervals of checkpointing an application is important for minimizing the run time of the application in the presence of system failures. Most of the existing efforts on checkpointing interval selection were developed for sequential applications while few efforts deal with parallel applications where the applications are executed on the same number of processors for the entire...

With the evolution of high-performance computing towards heterogeneous, massively parallel systems, parallel applications have developed new checkpoint and restart necessities. Whether due to a failure in the execution or to a migration of the application processes to different machines, checkpointing tools must be able to operate in heterogeneous environments. However, some of the data manipul...