Techniques for Designing Survivable Optical Grid Networks

Grid computing involves high performance computing with resource sharing to support data-intensive applications, and requires high speed communications. Wavelength division multiplexing (WDM) optical networks become a natural choice for interconnecting the distributed computational and/or storage resources due to their high throughput, high reliability and low cost. This has led to increased research attention on techniques for developing fault tolerant optical grids. Different solutions have been proposed in the literature to provide protection and fault-tolerance in both grid computing and optical networks. However, the design of a resilient optical grid network should consider the inter-relation between computing and networking resource usage and how they are affected by potential faults. Hence, it is necessary to develop integrated, sophisticated algorithms that jointly allocate both computing and networking resources to improve the resource utilization while guaranteeing service availability. In this paper, we investigate the survivability of optical grids and review the state-of-the-art techniques and approaches in this area. We also identify some open problems and point out a number of promising approaches for future research direction to achieve fault tolerance in optical grid networks.