Advance bandwidth scheduling algorithms in dedicated networks

Abstract

An increasing number of high-performance research network testbeds and production networks have the capability of provisioning dedicated channels for high-speed data transfer in support of large-scale scientific applications. Each dedicated channel in these networks typically consists of one or more physical links that are shared in both time and bandwidth through in-advance reservation among multiple applications. Efficiently allocating the available bandwidth of each component link across future time slots in response to an instant data transfer request is critical to maximizing network resource utilization and meeting diverse end-to-end transport performance requirements. Based on different path and bandwidth constraints, we formulate and investigate five bandwidth scheduling problems that aim to minimize the transfer end time for a given data size. For each of these scheduling problems, we design an optimal algorithm with polynomial-or pseudo-polynomial-time complexity with respect to the network size and total number of time slots in a bandwidth reservation table.