Resource scheduling heuristics for data intensive networks

摘要

In this work, we consider the adaptive capacity utilization problem for real-time heterogeneous data intensive MPLS (multiprotocol label switching) networks consisting of multiple source and destination nodes. We account for a file transfer scenario taking into account node capacity, file size, file transfer start time, and the deadline of files when determining the explicit capacity of virtual paths (VPs) across the backbone. We propose two heuristic algorithms referred to as the most conservative (MC) algorithm and the load balancing (LB) algorithm using a newly introduced parameter, capacity utilization. These heuristic algorithms are designed to react to file transfers according to the temporal capacity of VPs for multiple source and destination networks.