Download now Free registration required
Fault tolerance in parallel environments is a subject that has been studied extensively in theory, but practical implementations are relatively scarce. Clusters of commodity computers are commonly used to implement cost-effective, high-performance systems. However, the low-cost, commodity nature of clusters which leads to their principal drawback numerous low cost workstations are prone to the unexpected failures. It becomes critical to provide some degree of fault tolerance to the high performance applications running on these clusters.
- Format: PDF
- Size: 175.9 KB