Search

Congestion control protocols for background data are commonly conceived and designed to emulate low priority traffic, which yields to transmission control protocol (TCP) flows. In the presence of even a few very long TCP flows, this behavior can cause bandwidth starvation, and hence, the accumulation of large numbers of background data flows for prolonged periods of time, which may ultimately have an adverse effect on the download delays of delay-sensitive TCP flows. In this paper, we look at the fundamental problem of designing congestion control protocols for background traffic with the minimum impact on short TCP flows while achieving a certain desired average throughput over time . The corresponding optimal policy under various assumptions on the available information is obtained analytically. We give tight bounds of the distance between TCP-based background transfer protocols and the optimal policy, and identify the range of system parameters for which more sophisticated congestion control makes a noticeable difference. Based on these results, we propose an access control algorithm for systems where control on aggregates of background flows can be exercised, as in file servers. Simulations of simple network topologies suggest that this type of access control performs better than protocols emulating low priority over a wide range of parameters.

We consider the problem of designing access control protocols for servers distributing background data, such as software and database updates, so that they cause the least possible disruption to flows carrying delay-sensitive data. Using a Markov decision process formulation we obtain the optimal policy analytically, which is not easy to implement in practice. A mean-field argument is employed to show that another policy, which is easier to implement and is based on water-filling, converges to the optimal as the number of bottleneck links increases. Using simulations we compare the performance of this policy with the standard case where no control is exercised by the server.

We consider the design of transport protocols for background transfers with the aim of reducing the download delays of interactive (non-background) flows which share the same links. We demonstrate that Lower than Best Effort (LBE) protocols suffer from bandwidth starvation, a fact which not only creates adoption disincentives but can have a harmful effect on the interactive flows as well. Instead we propose a simple access control policy which limits the maximum number of simultaneously transmitting background flows. This maximum limit is dynamically adjusted in order to prevent bandwidth starvation thus eliminating the disincentives present in LBE protocols. Through simulations we show that our policy creates lower transfer delay for the interactive flows, as compared to that under LEDBAT [6] and TCP-LP [8] LBE protocols.