“Optimizing Deadline-Driven Bulk Data Transfers in Overlay Networks”
in 2009 Proceedings of 18th International Conference on Computer Communications and Networks, Aug. 2009.
Abstract: Deadline-driven bulk data transfers frequently occur in overlay networks running data-intensive, distributed work?ow applications, such as grid and cloud environments. What distinguishes such transfers from other Internet traffic is that overlay nodes should cooperate towards the common goal of delivering all interdependent data timely, rather than follow individual, selfish goals. For such scenarios, we propose scheduling transfers in overlays in a globally optimal manner with respect to minimizing overall network congestion (i.e. minimizing maximum link utilization). The challenge in this optimization is to jointly address routing of transfers within the overlay and time-domain transfer scheduling. We formally define and address the associated problem, the Bulk Data Routing and Transfer (BDRT). We present a linear programming-based solution to BDRT, optimal in both routing and time domains. We additionally explore alternative approaches based on heuristic routing strategies, both oblivious and time-domain optimized. We evaluate these solutions via both simulations and Internet transfer experiments. Our trace-driven simulations leverage PlanetLab public traces collected by the S3 measurement project, spanning about seven months. Measurement-based experiments are performed on the Intrigger wide-area grid and PlanetLab. Evaluation shows that our approach finds optimal solutions, based on estimations of job arrival times, deadlines and transfer volumes.
Research line(s): Timeliness and Adaptation in Dependable Systems (TADS)