Utility maximizing load balancing policies

TL;DR

This paper develops and analyzes load balancing policies that maximize utility in large-scale heterogeneous server systems, providing bounds and characterizations of their performance.

Contribution

It introduces two asymptotically optimal load balancing policies that achieve the maximum utility bound in large-scale systems.

Findings

Derived an upper bound for mean normalized utility in stationarity.

Proposed two load balancing policies that attain this upper bound.

Characterized the transient and stationary behavior of these policies.

Abstract

Consider a service system where incoming tasks are instantaneously dispatched to one out of many heterogeneous server pools. Associated with each server pool is a concave utility function which depends on the class of the server pool and its current occupancy. We derive an upper bound for the mean normalized aggregate utility in stationarity and introduce two load balancing policies that achieve this upper bound in a large-scale regime. Furthermore, the transient and stationary behavior of these asymptotically optimal load balancing policies is characterized on the scale of the number of server pools, in the same large-scale regime.