TCP-LP: Low-Priority Service via End-Point Congestion Control

Transmission Control Protocol-Low Priority (TCP-LP), a algorithm goal is to utilize only the excess network bandwidth as comparedto the “fair share” of bandwidth as targeted by TCP. Service prioritization among different traffic classes is an important goal for the Internet. Conventional approaches in solving this problem, considers the existing best-effort class as the low-priority class, and attempt to developmechanisms that provide “better-than-best-effort” service. We explore the opposite approach, and devise a new distributed algorithm to realize alow-priority service (as compared to the existing best effort) from the network endpoints. To this end, we develop TCP Low Priority (TCP-LP), a distributed algorithm whose goal is to utilize only the excess network bandwidth as compared to the “fair share” of bandwidth as targeted byTCP. The key mechanisms unique to TCP-LP congestion control are the use of one-way packet delays for early congestion indications and aTCP-transparent congestion avoidance policy. The results of our simulation and Internet experiments show that TCP-LP is largely non-intrusiveto TCP traffic. Both single and aggregate TCP-LP flows are able to successfully utilize excess network bandwidth; moreover, multiple TCP-LPflows share excess bandwidth fairly. Substantial amounts of excess bandwidth are available to the low-priority class, even in the presence of “greedy” TCP flows. Despite their low-priority nature, TCP-LP flows are able to utilize significant amounts of available bandwidth in a wideareanetwork environment