Distributed Server Allocation for Content Delivery Networks

TL;DR

This paper introduces a dynamic, cost-based model for server allocation in content delivery networks, proposing an index policy inspired by Whittle's approach and comparing its performance through simulations.

Contribution

It develops a novel Markov decision process formulation and a Whittle-like index policy for efficient server allocation in content delivery networks.

Findings

The index policy performs favorably compared to heuristics in simulations.

The model effectively balances cost and performance in server allocation.

The approach provides a scalable solution for dynamic content delivery networks.

Abstract

We propose a dynamic formulation of file-sharing networks in terms of an average cost Markov decision process with constraints. By analyzing a Whittle-like relaxation thereof, we propose an index policy in the spirit of Whittle and compare it by simulations with other natural heuristics.