Designing Low-Complexity Heavy-Traffic Delay-Optimal Load Balancing Schemes: Theory to Algorithms
Abstract
We establish a unified analytical framework for load balancing systems, which allows us to construct a general class of policies that are both throughput optimal and heavy-traffic delay optimal. This general class includes as special cases popular policies such as join-shortest-queue and power-of-, but not the join-idle-queue (JIQ) policy. In fact, we show that JIQ, which is not in , is actually not heavy-traffic delay optimal. Owing to the significant flexibility offered by class , we are able to design a new policy called join-below-threshold (JBT-d), which maintains the simplicity of pull-based policies such as JIQ, but updates its threshold dynamically. We prove that JBT- belongs to the class when the threshold is picked appropriately and thus it is heavy-traffic delay optimal. Extensive simulations show that the new policy not only has a low complexity in message rates, but also achieves excellent delay performance, comparable to the optimal join-shortest-queue in various system settings.
Cite
@article{arxiv.1710.04357,
title = {Designing Low-Complexity Heavy-Traffic Delay-Optimal Load Balancing Schemes: Theory to Algorithms},
author = {Xingyu Zhou and Fei Wu and Jian Tan and Yin Sun and Ness Shroff},
journal= {arXiv preprint arXiv:1710.04357},
year = {2017}
}
Comments
42 pages, 34 figure, this is the technical report for a paper accepted by ACM Sigmetrics 2018