Hide metadata

dc.date.accessioned2018-03-19T10:21:43Z
dc.date.available2018-03-19T10:21:43Z
dc.date.issued2017
dc.identifier.urihttp://hdl.handle.net/10852/61103
dc.description.abstractThe advent of the Internet of Things, sensor and social networks, to mention just a few examples, all contribute towards the solid establishment of the Big Data era. High Performance Computing (HPC) becomes necessary for the efficient processing of the massive amounts of data our society generates, and cloud computing is a critical component to deliver this processing power to a broader audience that cannot afford to acquire and maintain such complex computing systems themselves. However, HPC specific technology and performance is not yet apt to be delivered efficiently over highly flexible and dynamic environments, as typically are the virtualized cloud infrastructures. In this thesis, we address challenges that arise in high performance dynamic cloud environments that are equipped with HPC specific technology, in the context of networking and virtualization. We use InfiniBand, a high performance lossless interconnection network as the basis of our research, and first show that lossless networks pose prime challenges when the nature of the infrastructure is very dynamic, i.e. exhibits continuous changes. Then we propose a network I/O virtualization architecture, the InfiniBand vSwitch architecture, that can make lossless network technologies more favorable in the cloud. Moreover, we propose different network reconfiguration methods to enable performance-driven reconfigurations in very large network topologies that are commonly found in data centers. Performance-driven reconfigurations are frequently needed to adapt to unpredictable workload changes resulting from the shared and on-demand nature of a cloud platform, or when cloud providers employ live migration of virtual machines to optimize the resource usage of their infrastructure. Last but not least, we propose a new Quality-of-Service metric, called 'delay', to capture the directly observable service degradation in consolidated cloud environments. We suggest that the 'delay' can be used as a direct service level agreement metric between cloud providers and cloud tenants.en_US
dc.language.isoenen_US
dc.relation.haspartPaper I: A Novel Query Caching Scheme for Dynamic InfiniBand Subnets. Evangelos Tasoulas, Ernst Gunnar Gran, Bjørn Dag Johnsen, Tor Skeie. Published at IEEE/ACM International Symposium on Cluster, Cloud and Grid. Computing (CCGrid) 2015. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1109/CCGrid.2015.10
dc.relation.haspartPaper II: Towards the InfiniBand SR-IOV Architecture. Evangelos Tasoulas, Ernst Gunnar Gran, Bjørn Dag Johnsen, Kyrre Begnum, Tor Skeie. Published at IEEE International Conference on Cluster Computing (IEEECluster) 2015. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1109/CLUSTER.2015.58
dc.relation.haspartPaper III: Fast Hybrid Network Reconfiguration for Large-Scale Lossless Interconnection Networks. Evangelos Tasoulas, Ernst Gunnar Gran, Tor Skeie, Bjørn Dag Johnsen. Published at IEEE International Symposium on Network Computing and Applications (NCA) 2016. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1109/NCA.2016.7778601
dc.relation.haspartPaper IV: Compact Network Reconfiguration in Fat-Trees [138] Feroz Zahid, Ernst Gunnar Gran, Bartosz Bogdanski, Bjørn Dag Johnsen, Tor Skeie, Evangelos Tasoulas. Published at Springer Journal of Supercomputing in 2016, 72:4438-4467. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1007/s11227-016-1759-y
dc.relation.haspartPaper V: Efficient Routing and Reconfiguration in Virtualized HPC Environments with vSwitch-enabled Lossless Networks. Authors Evangelos Tasoulas, Feroz Zahid, Ernst Gunnar Gran, Kyrre Begnum, Bjørn Dag Johnsen, Tor Skeie. Published at Journal of Concurrency and Computation: Practice and Experience, 2018. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1002/cpe.4443
dc.relation.haspartPaper VI: The Concept of Workload Delay as a Quality-of-Service Metric for Consolidated Cloud Environments with Deadline Requirements. Evangelos Tasoulas, Hugo Lewi Hammer, Hårek Haugerud, Anis Yazidi, Alfred Bratterud, Boning Feng. Published at IEEE International Symposium on Network Computing and Applications (NCA) 2017. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1109/NCA.2017.8171342
dc.relation.hasparthttps://doi.org/10.1002/cpe.4443
dc.relation.urihttps://doi.org/10.1109/CCGrid.2015.10
dc.relation.urihttps://doi.org/10.1109/CLUSTER.2015.58
dc.relation.urihttps://doi.org/10.1109/NCA.2016.7778601
dc.relation.urihttps://doi.org/10.1007/s11227-016-1759-y
dc.relation.urihttps://doi.org/10.1109/NCA.2017.8171342
dc.titleTowards High Performance Dynamic Cloud Environmentsen_US
dc.typeDoctoral thesisen_US
dc.creator.authorTasoulas, Evangelos
dc.identifier.urnURN:NBN:no-63726
dc.type.documentDoktoravhandlingen_US
dc.identifier.fulltextFulltext https://www.duo.uio.no/bitstream/handle/10852/61103/1/Tasoulas-PhD-Thesis-2017.pdf


Files in this item

Appears in the following Collection

Hide metadata