Replica Selection and Downloading based on Wavelength Availability in Lambda-grid Networks
Grid computing integrates geographically distributed computing resources, such as CPU and storage, through communication networks. A data grid, one of the grid computing techniques, manages distributed data and offers high performance computing. This paper shows a typical model of a data grid network, which consists of many sites. A single site comprises a master node, a storage node, and a computing node. The master node schedules jobs and exchanges information such as the amount of free space of data storage.