Characteristics of Destination Address Locality in Computer Networks: A Comparison of Caching Schemes

TL;DR

This paper investigates the locality of destination address references in large computer networks and compares caching schemes to optimize packet forwarding speed, revealing different behaviors for interactive and noninteractive traffic.

Contribution

It provides an analysis of address reference locality and evaluates cache replacement algorithms, highlighting conditions where caching improves or hinders performance.

Findings

Interactive traffic shows different locality patterns than noninteractive traffic.

LRU caching performs well with noninteractive traffic but not with interactive traffic.

Caching effectiveness depends on cache size and traffic type.

Abstract

The size of computer networks, along with their bandwidths, is growing exponentially. To support these large, high-speed networks, it is neccessary to be able to forward packets in a few microseconds. One part of the forwarding operation consists of searching through a large address databse. This problem is encountered in the design of bridges, routers, gateways and name servers. Caching can reduce the lookup time if there is a locality in the address reference pattern. Using a destination reference trace measured on an extended local are a network, we attempt to see if the destination refernces do have a significant locality. We compared the performance of MIN, LRU, FIFO, and random cache replacement algorithms. We found that the interactive (terminal) traffic in our sample had quite different locality behavior than that of the noninteractive traffic. The interactive traffic did not follow the LRU stack model while the noninteractivetraffic did. Examples are shown of the environments in which caching can help as well as those in which caching can hurt, unless the cache size is large.