Reference : Load-Aware Strategies for Cloud-Based VoIP Optimization with VM Startup Prediction
Scientific congresses, symposiums and conference proceedings : Paper published in a book
Engineering, computing & technology : Computer science
http://hdl.handle.net/10993/34043
Load-Aware Strategies for Cloud-Based VoIP Optimization with VM Startup Prediction
English
Cortes-Mendoza, Jorge M. [> >]
Tchernykh, Andrei [> >]
Feoktistov, Alexander [> >]
Bouvry, Pascal mailto [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Computer Science and Communications Research Unit (CSC) >]
Didelot, Loïc [> >]
Jun-2017
IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), 2017
IEEE
Yes
International
IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), 2017
29 May-2 June, 2017
[en] Cloud computing, Quality of service, Codecs, Servers, Load modeling, Telephone sets, Bandwidth
[en] In this paper, we address cloud VoIP scheduling strategies to provide appropriate levels of quality of service to users, and cost to VoIP service providers. This bi-objective focus is reasonable and representative for real installations and applications. We conduct comprehensive simulation on real data of twenty three on-line non-clairvoyant scheduling strategies with fixed threshold of utilization to request VMs, and twenty strategies with dynamic prediction of the load. We show that our load-aware with predictions strategies outperform the known ones providing suitable quality of service and lower cost. The robustness of these strategies is also analyzed varying VM startup time delays to deal with realistic VoIP cloud environments.
http://hdl.handle.net/10993/34043
10.1109/IPDPSW.2017.73
http://ieeexplore.ieee.org/document/7965084/

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Limited access
PDCO2017VoIP.pdfPublisher postprint361.07 kBRequest a copy

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.