PROJECT TITLE :

Congestion Control for Background Data Transfers With Minimal Delay Impact - 2017

ABSTRACT:

Congestion management protocols for background data are commonly conceived and designed to emulate low priority traffic, that yields to transmission management protocol (TCP) flows. Within the presence of even some terribly long TCP flows, this behavior can cause bandwidth starvation, and hence, the buildup of large numbers of background information flows for prolonged periods of time, that could ultimately have an adverse impact on the download delays of delay-sensitive TCP flows. During this paper, we tend to look at the fundamental problem of planning 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 below numerous assumptions on the accessible info is obtained analytically. We give tight bounds of the distance between TCP-primarily based background transfer protocols and the optimal policy, and determine the vary of system parameters for which additional refined congestion control makes a noticeable difference. Based mostly on these results, we tend to propose an access management algorithm for systems where control on aggregates of background flows will be exercised, as in file servers. Simulations of straightforward network topologies suggest that this type of access control performs better than protocols emulating low priority over a wide selection of parameters.