Skip to Main content Skip to Navigation

Integrated optimization in cloud environment

Abstract : In geo-distributed cloud systems, a key challenge faced by cloud providers is to optimally tune and configure their underlying cloud infrastructure. An important problem in this context, deals with finding an optimal virtual machine (VM) placement, minimizing costs while at the same time ensuring good system performance. Moreover, due to the fluctuations of demand and traffic patterns, it is crucial to dynamically adjust the VM placement scheme over time. Hence, VM migration is used as a tool to cope with this problem. However, despite the benefits brought by VM migration, in geo-distributed cloud context, it generates additional traffic in the backbone links which may affect the application performance in both source and destination DCs. Hence, migration decisions need to be effective and based on accurate parameters. In this work, we study optimization problems related to the placement, migration and scheduling of VMs hosting highly correlated and distributed applications within geo-distributed DCs. In this context, we propose an autonomic DC management tool based on both online and offline optimization models to manage the distributed cloud infrastructure. Our objective is to minimize the overall expected traffic volume circulating between the different DCs of the system. To deal with different types of communication traffic patterns, we propose both deterministic and stochastic optimization models to solve VM placement and migration problem and to cope with the uncertainty of inter-VM traffic. Furthermore, we propose near-optimal algorithms that provide with the best inter-DCs migration sequence of inter-communicating VMs. Along with that, we study the impact of the VM's lifetime on the migration decisions in order to maintain the stability of the cloud system. Finally, to evaluate and validate our approach, we use experimental tests as well as simulation environments. The results of the conducted experiments show the effectiveness of our proposals
Complete list of metadatas

Cited literature [132 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Thursday, February 8, 2018 - 11:25:10 AM
Last modification on : Wednesday, June 24, 2020 - 4:18:24 PM
Long-term archiving on: : Saturday, May 5, 2018 - 4:46:33 AM


Version validated by the jury (STAR)


  • HAL Id : tel-01704075, version 1


Hana Teyeb. Integrated optimization in cloud environment. Networking and Internet Architecture [cs.NI]. Université Paris-Saclay; Université Tunis El Manar. Faculté des Sciences Mathématiques, Physiques et Naturelles de Tunis (Tunisie), 2017. English. ⟨NNT : 2017SACLL010⟩. ⟨tel-01704075⟩



Record views


Files downloads