Each cloud infrastructure-as-a-service (IaaS) provider offers its own set of virtual machine (VM) images and hypervisors. This creates a vendor lock-in problem when cloud users try to change cloud provider (CP). Although, recently a few user-side inter-cloud migration techniques have been proposed (e.g. nested virtualisation), these techniques do not provide dynamic cloud management facilities which could help users to decide whether or not to proceed with migration, when and where to migrate, etc. Such decision-making support in the post-deployment phase is crucial when the current CP's Quality of Service (QoS) degrades while other CPs offer better QoS or the same service at a lower price. To ensure that users' required QoS constraints are achieved, dynamic monitoring and management of the acquired cloud services are very important and should be integrated with the inter-cloud migration techniques. In this paper, we present the problem formulation and the architecture of a Multi-objective dYnamic MIgratioN Decision makER (MyMinder) framework that enables users to monitor and appropriately manage their deployed applications by providing decisions on whether to continue with the currently selected CP or to migrate to a different CP. The paper also discusses experimental results obtained when running a Spark linear regression application in Amazon EC2 and Microsoft Azure as an initial investigation to understand the motivating factors for live-migration of cloud applications across cloud providers in the post-deployment phase.
|Title of host publication||Proceeding of the 7th International Conference on Cloud Computing and Services Science|
|Editors||Donald Ferguson, Victor Mendez Munoz, Jorge Cardoso, Markus Helfert, Claus Pahl|
|Number of pages||8|
|Publication status||Published - 26 Apr 2017|
Barlaskar, E., Kilpatrick, P., Spence, I., & Nikolopoulos, D. S. (2017). MyMinder: A User-centric Decision Making Framework for Intercloud Migration. In D. Ferguson, V. Mendez Munoz, J. Cardoso, M. Helfert, & C. Pahl (Eds.), Proceeding of the 7th International Conference on Cloud Computing and Services Science (pp. 588-595). SciTePress. https://doi.org/10.5220/0006355905880595