On delays in management frameworks: Metrics, models and analysis

2Citations
Citations of this article
3Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

Management performance evaluation means assessment of scalability, complexity, accuracy, throughput, delays and resources consumptions. In this paper, we focus on the evaluation of management frameworks delays through a set of specific metrics. We investigate the statistical properties of these metrics when the number of management nodes increases. We show that management delays measured at the application level are statistically modeled by distributions with heavy tails, especially the Weibull distribution. Given that delays can substantially degrade the capacity of management algorithms to react and resolve problems it is useful to get a finer model to describe them. We suggest the Weibull distribution as a model of delays for the analysis and simulations of such algorithms. © IFIP International Federation for Information Processing 2006.

Cite

CITATION STYLE

APA

Lahmadi, A., Andrey, L., & Festor, O. (2006). On delays in management frameworks: Metrics, models and analysis. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4269 LNCS, pp. 13–24). Springer Verlag. https://doi.org/10.1007/11907466_2

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free