Strategies for fault tolerance in optical grid networks

The need for powerful computing resources as well as capabilities for storage and transmission of large amounts of data in a number of application areas have led to the emergence of optical grids as a natural, cost-effective platform for supporting such applications. As a result there is also an increasing need for strategies and techniques designed to achieve fault tolerance in optical grid networks. Design for fault tolerance in both grid computing and optical networks are mature, well-researched fields in their own right. However, survivability in optical grids should not be treated merely as a concatenation of techniques developed separately in these two disciplines. Rather, it would be beneficial, in terms of resource availability as well as cost-effectiveness, to develop an integrated approach that takes into consideration the allocation of both computing and networking resources jointly. In this paper, we review the state-of-the-art techniques and approaches that have been proposed in the literature, for designing survivable optical grid networks. We also discuss some challenges, identify some open problems and outline future research directions for developing an integrated approach to fault tolerance in optical grids.