Maximum Likelihood Estimation of Closed Queueing Network Demands from Queue Length Data

Abstract

Resource demand estimation is essential for the application of analyical models, such as queueing networks, to realworld systems. In this paper, we investigate maximum likelihood (ML) estimators for service demands in closed queueing networks with load-independent and load-dependent service times. Stemming from a characterization of necessary conditions for ML estimation, we propose new estimators that infer demands from queue-length measurements, which are inexpensive metrics to collect in real systems. One advantage of focusing on queue-length data compared to response times or utilizations is that confidence intervals can be rigorously derived from the equilibrium distribution of the queueing network model. Our estimators and their confidence intervals are validated against simulation and real system measurements for a multi-tier application.

Performance Modelling and Optimisation of NoSQL Database Systems

“…In addition, [KKRD12] introduces the concept of confidence in the demand estimation with this method. Differently, [WCKN16] propose the QMLE algorithm that uses mean queue length values, rather than response times, to perform demand estimation. The queue-length is also used by other algorithms such as Gibbs sampling and Bayesian inference [SJ11,WC13].…”

SD: A Divergence-Based Estimation Method for Service Demands in Cloud Systems

“…Another machine learning techniques for demand inference includes clustering [15], which starting from observation data composed by timestamps, throughput and utilization, can recognize deviations over time of demands, such as those resulting from hardware upgrades. Recently, [16] propose the QMLE algorithm, a technique based on the maximum likelihood estimation, that uses mean queue length values, rather than response times, to perform demand estimation.…”