TCP-Cherry for satellite IP networks: Analytical model and performance evaluation

Abstract

TCP-Cherry is a novel TCP congestion control scheme that we devised for ensuring high performance over satellite IP networks and the alikes which are characterized by long propagation delays and high link errors. In TCP-Cherry, two new algorithms, Fast-Forward Start and First-Aid Recovery, have been proposed for congestion control. Our algorithms use supplement segments, i.e., low-priority segments to probe the available bandwidth in the network for the TCP connections along with carrying new data blocks. In this paper, we present our new congestion control scheme, TCP-Cherry and devise an analytical model for it. Our major contributions in this paper include the analytical model and equations for performance evaluation, validation of the analytical model through comparison between analytical and simulation results and devising a guideline to tune the buffer related parameters both at the sender as well as the receiver ends for optimum throughput performance. Experiments show that simulation results and the calculated throughput from our analytical model match quite closely, thereby verifying the appropriateness of the model. In addition, from analysis of simulation results, we discover that a buffer size at the receiver, rwnd, that is around four times maxcwnd, or the maximum congestion window at the sender side, is likely to maintain high throughput over a wide range of operating conditions.

Introduction

Conventional TCP protocols suffer from performance problems over satellite links that are characterized by long propagation delay and relatively high link error rates [1], [2]. Since TCP was originally designed for wired networks with almost no link error, occurrence of link error is misinterpreted as congestion by normal TCP. This ultimately leads to performance degradation [3], [4].

Several approaches toward solving these problems exist. Among them, TCP-Peach [1] and TCP-Peach+ [5] worth special attention. TCP-Peach [1] is an innovative congestion control scheme that deploys a bandwidth measurement mechanism based on low-priority segments. TCP-Peach+ [5] extends the idea further. Another scheme, TCP-Hybla [6], performs even better on long RTT links like satellite links over a wide range of link error rates.

In [7], we propose a new approach to TCP congestion control over satellite links. We name this as TCP-Cherry. TCP-Cherry takes into account some inherent causes of TCP performance degradation which the existing schemes do not take into consideration. The basic contribution of TCP-Cherry consists of the introduction of data carrying low priority segments, called supplement segments; the algorithms Fast-Forward Start and First-Aid Recovery – the counterparts of conventional TCP Slow Start and Fast Recovery, respectively; the mechanism to avoid irreversible gradual decrease of TCP congestion window owing to recurring loss of probing segments and the mechanism for selection of supplement segments which keeps the overhead of duplicate transmission minimum. Simulation results show that TCP-Cherry yields a performance betterment of more than 150% compared with the nearest counterpart, TCP-Hybla [7].

In this paper, we present our new scheme TCP-Cherry and propose an analytical model for it. In addition, we investigate the impacts of window size parameters and provide guidelines for maintaining high performance while deploying TCP-Cherry.

Our major contributions in this paper are (i) introducing the analytical model for TCP-Cherry; the analytical equations Eqs. (6), (7), (8), (9), (10), (11), (ii) evaluating TCP-Cherry throughput with the model, (iv) evaluating TCP-Cherry throughput through simulation for validating of the analytical model, and (v) from simulations of TCP-Cherry developing a guideline for selection of optimal parameters of TCP-Cherry for efficient operations over a spectrum of operating conditions.

The rest of this paper is organized as follows. At first, we explain TCP-Cherry in Section 2. Next we introduce its analytical model in Section 3. In Section 4 we evaluate the performance of the new scheme through analysis as well as simulation. In Section 5 we discuss about optimal parameters of TCP-Cherry from the simulation results. Finally, in Section 6, we conclude the paper.