Weibull mixture model to characterise end-to-end Internet delay at coarse time-scales

Abstract

Traces collected at monitored points around the Internet contain representative performance information about the paths their probes traverse. If processed appropriately, basic measurement attributes, such as delay and loss, can be used to output conclusions about the performance status of the network, with subsequent applications in fault and performance management, network provisioning, traffic engineering and performance prediction. However, the task of analysis and extracting such valuable information from measurements only remains challenging. The Weibull mixture model, a method to characterise end-to-end network delay measurements within a few simple, accurate, representative and handleable parameters using a finite combination of Weibull distributions is presented. The model parameters are related to meaningful delay characteristics, such as average peak and tail behaviour in a daily profile, and can be optimally found using an iterative algorithm known as expectation maximisation. Studies on such parameter evolution can reflect current workload status and all possible network events impacting packet dynamics, with further applications in network management. The model is further tested and validated with real GPS synchronised measurements taken across the Internet, donated by RIPE NCC.