Optimizing End-to-End Throughput for Data Transfers on an Overlay-TCP Path

Abstract

Abstract.We consider an overlay network where traffic on an overlay hop is carried in one or more TCP connections established between the overlay nodes at the ends of the hop. We are interested in maximizing the throughput of data carried by this type of overlay network. To that end, we focus on a single path in such a network and investigate how one can determine the number of TCP connections on each overlay hop so as to maximize the throughput of the data carried end-to-end on that path. We first show that having more than one TCP connection on some overlay hops can indeed increase the throughput on overlay paths. We then propose the Adaptive Overlay-TCP Provisioning approach, that, based on the path state, dynamically introduces and removes TCP connections on individual overlay hops to maximize throughput while minimizing the overhead of extraneous connections. We consider two schemes to assess the path state, the intermediate buffer occupancy scheme and the isolated rate probing scheme, and evaluate them experimentally on a set of Planetlab nodes. We show that these schemes can significantly improve the end-to-end throughput with very little overhead.

Fair Allocation of Substrate Resources among Multiple Overlay Networks

“…We show that these definitions are also applicable in the multiple overlay network setting. The authors also worked on maximizing the throughput of end-to-end data carried on overlay paths using TCP [8], but they considered a single overlay network rather than multiple overlays competing for resources, as in the scenario in this paper. This problem is similar to the virtual network assignment (VNA) problem [13] in that both problems are concerned with the efficient allocation of substrate resources to overlay/virtual networks.…”

Optimizing Deadline-Driven Bulk Data Transfers in Overlay Networks

“…To date, we have not found an evaluation of oblivious algorithms against an optimal solution through either simulations or experiments, at least not in the context of actual Internet overlays. By tackling the overlay routing problem, we also relate to a plethora of work on routing overlay networks, such as [1][11] [7]. Space limits do not allow for an overview of this area.…”

“…In the context of a besteffort Internet, unable to provide QoS guarantees for applications, systems such as opportunistic routing overlays [1][11] [7] have emerged over the past decade to provide a flexible and interesting way to get extra QoS out of public networks by only manipulating the application level, with no cooperation from the underlying network. Such overlays typically attempt to improve QoS by routing around congestion and network bottlenecks.…”