Parallel TCP Flows Research Articles

An Extension of the TCP Steady-State Throughput Equation for Parallel Flows and Its Application in MulTFRC

In the first part of this paper, we present a simple extension of the well-known TCP steady-state throughput equation that can be used to calculate the throughput of several flows that share an end-to-end path. The value of this extension, which we show to work well with simulations as well as real-life measurements, is its practical applicability. Thus, in the second part of this paper, we present its application in MulTFRC, a TCP-friendly rate control (TFRC)-based congestion control mechanism that is fair to a number of parallel TCP flows while maintaining a smoother sending rate than multiple real TFRC flows do. MulTFRC enables its users to prioritize transfers by controlling the fairness among them in an almost arbitrary fashion.

Adjusting the Aggregate Throughput of Parallel TCP Flows without Central Coordination

We propose an Adjustable Parallel TCP (AP-TCP) which is a new scheme to control the aggregate throughput of parallel TCP flows. The AP-TCP can adjust the aggregate throughput to be any desired level irrespective of the parallel size (the number of parallel TCP flows). To adjust the aggregate throughput, we modify the increment factor of each parallel TCP flow to K 2 /N 2 where N is the number of parallel TCP flows and K is a value equivalent to any desired level for the aggregate throughput. Once K is given, the AP-TCP attempts to have K times more bandwidth than a single TCP flow when they are competing on the same network path. Another feature of the AP-TCP is its self-adjustment scheme. There is no central coordination or control overhead for parallel TCP flows. We analyze the model of the AP-TCP theoretically and evaluate it by using NS-2 simulation.

A Receiver-Driven Transport Protocol for the Web

In this paper, we present a receiver-oriented, request/response protocol for the Web that is compatible with the dynamics of TCP's congestion control algorithm. The protocol, called WebTP, is designed to be completely receiver-based in terms of transport initiation, flow-control and congestion-control. We propose a dual window-cum-rate based congestion control mechanism that is compatible with parallel TCP flows, and in fact interacts better with a congested network state. In support of our receiver-driven design, we developed a novel retransmission scheme that is robust to delay variations. The resulting flows achieve efficient network utilization and are qualitatively fair in their interaction amongst themselves and even with competing TCP flows. The paper also provides detailed simulation results to support the protocol design.