Abstract
Benefiting from the convenience of virtualization, virtual machine migration is generally utilized to fulfil optimization objectives in cloud/edge computing. However, live migration has certain risks and unapt decision may lead to side effects and performance degradation. Leveraging modified deep Q network, this paper provided an advanced risk evaluation system. Thorough formulation was given in this paper and a specific integration method was innovated based on uncertain theory. Series experiments were carried on computing cluster with OpenStack. The experimental results showed deep Q network for risk system was reliable while the uncertain approach was a proper way to deal with the risk integration.
Highlights
Cloud computing has become a consolidated computing paradigm, which allows users around the world to submit various computing requests
The source physical machine (PM), the destination PM, the target virtual machine (VM), the resource status along the route path, and the migration opportunity, all these factors are related to the optimal solution for live VM migration
Bionics algorithms are more refined; Scientific Programming they cannot cope with the frequently evolving environment in cloud computing. Different from these approaches, focusing on the live migration decision issue, this paper provides a specific risk evaluation method based on deep Q network and the uncertain theory. e unique highlight of this paper is to investigate the migration decision issue from AI perspective
Summary
Cloud computing has become a consolidated computing paradigm, which allows users around the world to submit various computing requests. Due to the high complexity of heterogeneous computing cluster, heuristic algorithms with fixed migration threshold for CPU/memory utilization are usually applied to deal with primary optimization goals (e.g., load balance, energy conservation) in live VM migration. Bionics algorithms are more refined; Scientific Programming they cannot cope with the frequently evolving environment (the resource configuration, the workload, and the requirement/constraint are constantly changing) in cloud computing. Different from these approaches, focusing on the live migration decision issue, this paper provides a specific risk evaluation method based on deep Q network and the uncertain theory.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
