Congestion Avoidance in Computer Networks with a Connectionless Network Layer

TL;DR

This paper explores congestion avoidance in connectionless computer networks, proposing a binary feedback scheme that prevents congestion by adjusting load, demonstrating its efficiency, fairness, and robustness through simulation.

Contribution

Introduces a binary feedback congestion avoidance scheme for connectionless networks, emphasizing its global optimality, efficiency, and fairness, with simulation validation.

Findings

The scheme is globally efficient and fair.

It is responsive, convergent, and robust.

The approach is configuration-independent.

Abstract

Widespread use of computer networks and the use of varied technology for the interconnection of computers has made congestion a significant problem. In this report, we summarize our research on congestion avoidance. We compare the concept of congestion avoidance with that of congestion control. Briefly, congestion control is a recovery mechanism, while congestion avoidance is a prevention mechanism. A congestion control scheme helps the network to recover from the congestion state while a congestion avoidance scheme allows a network to operate in the region of low delay and high throughput with minimal queuing, thereby preventing it from entering the congested state in which packets are lost due to buffer shortage. A number of possible alternatives for congestion avoidance were identified. From these alternatives we selected one called the binary feedback scheme in which the network uses a single bit in the network layer header to feed back the congestion information to its users, which then increase or decrease their load to make optimal use of the resources. The concept of global optimality in a distributed system is defined in terms of efficiency and fairness such that they can be independently quantified and apply to any number of resources and users. The proposed scheme has been simulated and shown to be globally efficient, fair, responsive, convergent, robust, distributed, and configuration-independent.