Aggregated service consolidation by multi-server virtualization

Simple item page
dc.contributor.advisorMurgu, Alexandruen_ZA
dc.contributor.authorNkubito, Phillip Uwizeyeen_ZA
dc.date.accessioned2014-07-31T10:54:15Z
dc.date.available2014-07-31T10:54:15Z
dc.date.issued2012en_ZA
dc.descriptionIncludes abstract.
dc.descriptionIncludes bibliographical references.
dc.description.abstractVirtualization is seen as the underlying technology that will make ubiquitous internet a reality by giving service providers a means to manage and consolidate their services on fewer infrastructures, while also being able to still deliver agreed service levels to their customers. Data centres have taken on server virtualization due to its benefits such as reduced operational cost, centralized administration, enhanced security, load balancing and improved hardware utilization. The resource management capabilities offered by server virtualization implies that these providers can now make use of previously redundant computing resources by aggregating them into a pool of resources. Virtual machine (VM) live migration is a core feature in resource management as it allows running applications in the VMs to be relocated to another host so as to maintain Service Level Agreements (SLAs) or for host maintenance purposes, with near zero degradation to the performance of the hosted service. However there is a need to manage optimally the live migration process especially when dealing with heterogeneous workloads such as CPU, memory and network intensive hosted applications which put a strain on different computing resources. By doing so, a service provider reaps maximum benefits from the live migration process. This study aims to optimize the live migration process during the migration of a VM running a Voice over IP (VoIP) service from a source to target host residing on the same Local Area Network. The proposed design framework focused on modifying the pre-copy algorithm which is in charge of performing the live migration. The modifications introduced a statistical dirty page tracking scheme into the algorithm and also incorporated the use of CPU scheduling to improve the performance of the live migration. A proof of concept was setup in the form of a testbed using Xen as the virtualization platform. Experiments carried out, showed that the downtime and migration time of the VoIP service were significantly reduced with minimal degradation on the call quality as perceived by the user during the live migration.en_ZA
dc.identifier.apacitationNkubito, P. U. (2012). <i>Aggregated service consolidation by multi-server virtualization</i>. (Thesis). University of Cape Town ,Faculty of Engineering & the Built Environment ,Department of Electrical Engineering. Retrieved from http://hdl.handle.net/11427/5136en_ZA
dc.identifier.chicagocitationNkubito, Phillip Uwizeye. <i>"Aggregated service consolidation by multi-server virtualization."</i> Thesis., University of Cape Town ,Faculty of Engineering & the Built Environment ,Department of Electrical Engineering, 2012. http://hdl.handle.net/11427/5136en_ZA
dc.identifier.citationNkubito, P. 2012. Aggregated service consolidation by multi-server virtualization. University of Cape Town.en_ZA
dc.identifier.ris TY - Thesis / Dissertation AU - Nkubito, Phillip Uwizeye AB - Virtualization is seen as the underlying technology that will make ubiquitous internet a reality by giving service providers a means to manage and consolidate their services on fewer infrastructures, while also being able to still deliver agreed service levels to their customers. Data centres have taken on server virtualization due to its benefits such as reduced operational cost, centralized administration, enhanced security, load balancing and improved hardware utilization. The resource management capabilities offered by server virtualization implies that these providers can now make use of previously redundant computing resources by aggregating them into a pool of resources. Virtual machine (VM) live migration is a core feature in resource management as it allows running applications in the VMs to be relocated to another host so as to maintain Service Level Agreements (SLAs) or for host maintenance purposes, with near zero degradation to the performance of the hosted service. However there is a need to manage optimally the live migration process especially when dealing with heterogeneous workloads such as CPU, memory and network intensive hosted applications which put a strain on different computing resources. By doing so, a service provider reaps maximum benefits from the live migration process. This study aims to optimize the live migration process during the migration of a VM running a Voice over IP (VoIP) service from a source to target host residing on the same Local Area Network. The proposed design framework focused on modifying the pre-copy algorithm which is in charge of performing the live migration. The modifications introduced a statistical dirty page tracking scheme into the algorithm and also incorporated the use of CPU scheduling to improve the performance of the live migration. A proof of concept was setup in the form of a testbed using Xen as the virtualization platform. Experiments carried out, showed that the downtime and migration time of the VoIP service were significantly reduced with minimal degradation on the call quality as perceived by the user during the live migration. DA - 2012 DB - OpenUCT DP - University of Cape Town LK - https://open.uct.ac.za PB - University of Cape Town PY - 2012 T1 - Aggregated service consolidation by multi-server virtualization TI - Aggregated service consolidation by multi-server virtualization UR - http://hdl.handle.net/11427/5136 ER - en_ZA
dc.identifier.urihttp://hdl.handle.net/11427/5136
dc.identifier.vancouvercitationNkubito PU. Aggregated service consolidation by multi-server virtualization. [Thesis]. University of Cape Town ,Faculty of Engineering & the Built Environment ,Department of Electrical Engineering, 2012 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/5136en_ZA
dc.language.isoengen_ZA
dc.publisher.departmentDepartment of Electrical Engineeringen_ZA
dc.publisher.facultyFaculty of Engineering and the Built Environment
dc.publisher.institutionUniversity of Cape Town
dc.subject.otherElectrical Engineeringen_ZA
dc.titleAggregated service consolidation by multi-server virtualizationen_ZA
dc.typeMaster Thesis
dc.type.qualificationlevelMasters
dc.type.qualificationnameMScen_ZA
uct.type.filetypeText
uct.type.filetypeImage
uct.type.publicationResearchen_ZA
uct.type.resourceThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
thesis_ebe_2012_nkubito_p.pdf
Size:
3.98 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Masters