Identifying Logical Homogeneous Clusters for Efficient Wide-area Communications

TL;DR

This paper proposes a method to identify logically homogeneous clusters in wide-area networks by analyzing network information, aiming to improve the modeling of communication performance in heterogeneous distributed systems.

Contribution

It introduces a novel strategy for constructing logical homogeneous clusters based on network data, enhancing the accuracy of performance modeling in wide-area environments.

Findings

Effective clustering of networks based on homogeneity improves communication modeling.

The approach outperforms traditional IP subnet-based clustering methods.

Enhanced performance predictions in wide-area distributed systems.

Abstract

Recently, many works focus on the implementation of collective communication operations adapted to wide area computational systems, like computational Grids or global-computing. Due to the inherently heterogeneity of such environments, most works separate "clusters" in different hierarchy levels. to better model the communication. However, in our opinion, such works do not give enough attention to the delimitation of such clusters, as they normally use the locality or the IP subnet from the machines to delimit a cluster without verifying the "homogeneity" of such clusters. In this paper, we describe a strategy to gather network information from different local-area networks and to construct "logical homogeneous clusters", better suited to the performance modelling.