The Multi-Branched Method of Moments for Queueing Networks

TL;DR

This paper introduces an enhanced Method of Moments algorithm for closed multiclass queueing networks, significantly increasing computational speed and reducing memory usage compared to traditional methods like MVA.

Contribution

The paper generalizes the Method of Moments by incorporating multiple recursive branches, resulting in a simpler matrix difference equation and substantial computational improvements.

Findings

Algorithm is 1,000 to 10,000 times faster than original MoM.

Reduces memory consumption significantly.

Extends feasible exact solution range for multiclass models.

Abstract

We propose a new exact solution algorithm for closed multiclass product-form queueing networks that is several orders of magnitude faster and less memory consuming than established methods for multiclass models, such as the Mean Value Analysis (MVA) algorithm. The technique is an important generalization of the recently proposed Method of Moments (MoM) which, differently from MVA, recursively computes higher-order moments of queue-lengths instead of mean values. The main contribution of this paper is to prove that the information used in the MoM recursion can be increased by considering multiple recursive branches that evaluate models with different number of queues. This reformulation allows to formulate a simpler matrix difference equation which leads to large computational savings with respect to the original MoM recursion. Computational analysis shows several cases where the proposed algorithm is between 1,000 and 10,000 times faster and less memory consuming than the original MoM, thus extending the range of multiclass models where exact solutions are feasible.