Robust Workload Estimation in Queueing Network Performance Models

Author(s):  
Giuliano Casale ◽  
Paolo Cremonesi ◽  
Roberto Turrin
2007 ◽  
Vol 44 (02) ◽  
pp. 321-331
Author(s):  
Heng-Qing Ye

In this paper we present counter-intuitive examples for the multiclass queueing network, where each station may serve more than one job class with differentiated service priority and each job may require service sequentially by more than one service station. In our examples, the network performance is improved even when more jobs are admitted for service.


2007 ◽  
Vol 44 (2) ◽  
pp. 321-331
Author(s):  
Heng-Qing Ye

In this paper we present counter-intuitive examples for the multiclass queueing network, where each station may serve more than one job class with differentiated service priority and each job may require service sequentially by more than one service station. In our examples, the network performance is improved even when more jobs are admitted for service.


2021 ◽  
Author(s):  
Yasir Shoaib

The performance characteristics such as throughput, resource utilization and response time of a system can be determined through measurement, simulation modeling and analytic modeling. In this thesis, measurement and analytic modeling approaches are applied to study the performance of a Apache-PHP-PostgreSQL web application. Layered Queueing Network (LQN) analytic modeling has been used to represent the system's performance model. The measurements found from load testing are compared with model analysis results for model validation. This thesis aims to show that LQN performance models are versatile enough to allow development of highly granular and easily modifiable models of PHP-based web applications and furthermore are capable of performance prediction with sufficiently high accuracy. Lastly, the thesis also describes utilities and methods used for load testing and determination of service demand parameters in our research work which would aid in shortening time required in development and study of performance models of similar systems.


2013 ◽  
Vol 50 (1) ◽  
pp. 151-165 ◽  
Author(s):  
Hendrik Baumann ◽  
Werner Sandmann

Stationary expectations corresponding to long-run averages of additive functionals on level-dependent quasi-birth-and-death processes are considered. Special cases include long-run average costs or rewards, moments and cumulants of steady-state queueing network performance measures, and many others. We provide a matrix-analytic scheme for numerically computing such stationary expectations without explicitly computing the stationary distribution of the process, which yields an algorithm that is as quick as its counterparts for stationary distributions but requires far less computer storage. Specific problems arising in the case of infinite state spaces are discussed and the application of the algorithm is demonstrated by a queueing network example.


2013 ◽  
Vol 50 (01) ◽  
pp. 151-165 ◽  
Author(s):  
Hendrik Baumann ◽  
Werner Sandmann

Stationary expectations corresponding to long-run averages of additive functionals on level-dependent quasi-birth-and-death processes are considered. Special cases include long-run average costs or rewards, moments and cumulants of steady-state queueing network performance measures, and many others. We provide a matrix-analytic scheme for numerically computing such stationary expectations without explicitly computing the stationary distribution of the process, which yields an algorithm that is as quick as its counterparts for stationary distributions but requires far less computer storage. Specific problems arising in the case of infinite state spaces are discussed and the application of the algorithm is demonstrated by a queueing network example.


2015 ◽  
Vol 52 (3) ◽  
pp. 609-621 ◽  
Author(s):  
Hendrik Baumann ◽  
Werner Sandmann

We consider long-run averages of additive functionals on infinite discrete-state Markov chains, either continuous or discrete in time. Special cases include long-run average costs or rewards, stationary moments of the components of ergodic multi-dimensional Markov chains, queueing network performance measures, and many others. By exploiting Foster-Lyapunov-type criteria involving drift conditions for the finiteness of long-run averages we determine suitable finite subsets of the state space such that the truncation error is bounded. Illustrative examples demonstrate the application of this method.


2015 ◽  
Vol 52 (03) ◽  
pp. 609-621 ◽  
Author(s):  
Hendrik Baumann ◽  
Werner Sandmann

We consider long-run averages of additive functionals on infinite discrete-state Markov chains, either continuous or discrete in time. Special cases include long-run average costs or rewards, stationary moments of the components of ergodic multi-dimensional Markov chains, queueing network performance measures, and many others. By exploiting Foster-Lyapunov-type criteria involving drift conditions for the finiteness of long-run averages we determine suitable finite subsets of the state space such that the truncation error is bounded. Illustrative examples demonstrate the application of this method.


Sign in / Sign up

Export Citation Format

Share Document