Network Load Balancing Methods: Experimental Comparisons and Improvement

TL;DR

This paper compares four load balancing algorithms through simulation, evaluates their performance, and proposes an improvement for the shortest queue method using stale load information.

Contribution

It provides an experimental comparison of load balancing methods and introduces an enhancement for the shortest queue algorithm with stale load data.

Findings

Shortest queue with stale load info reduces mean delay.

Round robin performs consistently across scenarios.

Improved shortest queue method shows significant performance gains.

Abstract

Load balancing algorithms play critical roles in systems where the workload has to be distributed across multiple resources, such as cores in multiprocessor system, computers in distributed computing, and network links. In this paper, we study and evaluate four load balancing methods: random, round robin, shortest queue, and shortest queue with stale load information. We build a simulation model and compare mean delay of the systems for the load balancing methods. We also provide a method to improve shortest queue with stale load information load balancing. A performance analysis for the improvement is also presented in this paper.