Towards a client-side QoS monitoring and assessment using Generalized Pareto Distribution in a cloud-based environment

Ammar Kamel, Ala Al-Fuqaha, Dionysios Kountanis, Issa Khalil

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

Recently, cloud computing and cloud-based services have been gaining a lot of momentum due to their cost effectiveness. Assuring Quality-of-Service (QoS) guarantees to the cloud's mobile clients is an important and unavoidable problem especially for delay sensitive application like VoIP. In this paper, we propose a novel approach for the early detection of cloud service degradations. The proposed algorithms employ the Generalized Pareto Distribution (GPD) approach to converge to a precise QoS model based on collected delay measurements from the cloud's mobile clients. Our experimental results demonstrate the usefulness of the proposed techniques and their ability to perform early detection of performance degradation of the cloud's infrastructure and its hosted services.

Original languageEnglish
Title of host publication2013 IEEE Wireless Communications and Networking Conference Workshops, WCNCW 2013
Pages123-128
Number of pages6
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event2013 IEEE Wireless Communications and Networking Conference Workshops, WCNCW 2013 - Shanghai, China
Duration: 7 Apr 201310 Apr 2013

Publication series

Name2013 IEEE Wireless Communications and Networking Conference Workshops, WCNCW 2013

Conference

Conference2013 IEEE Wireless Communications and Networking Conference Workshops, WCNCW 2013
Country/TerritoryChina
CityShanghai
Period7/04/1310/04/13

Keywords

  • Extreme Value Theorem
  • Extreme Values
  • Generalized Extreme Value
  • Generalized Pareto Distribution
  • Peak over Threshold
  • Quality of Service

Fingerprint

Dive into the research topics of 'Towards a client-side QoS monitoring and assessment using Generalized Pareto Distribution in a cloud-based environment'. Together they form a unique fingerprint.

Cite this