Live migration of virtual machines
NSDI'05 Proceedings of the 2nd conference on Symposium on Networked Systems Design & Implementation - Volume 2
Autonomic Live Adaptation of Virtual Computational Environments in a Multi-Domain Infrastructure
ICAC '06 Proceedings of the 2006 IEEE International Conference on Autonomic Computing
Sandpiper: Black-box and gray-box resource management for virtual machines
Computer Networks: The International Journal of Computer and Telecommunications Networking
Adaptive virtual network provisioning
Proceedings of the second ACM SIGCOMM workshop on Virtualized infrastructure systems and architectures
Non-intrusive virtualization management using libvirt
Proceedings of the Conference on Design, Automation and Test in Europe
Evaluating Xen-based virtual routers performance
International Journal of Communication Networks and Distributed Systems
Principles of Elastic Processes
IEEE Internet Computing
| Hi-index | 0.00 |
Cloud computing offers on-demand access to computational resources. One of the major challenges in cloud environments is to enforce the elasticity of the processes that execute in the cloud, avoiding Service Level Agreements (SLAs) violations and reducing waste with idle resources. We propose an autonomic resource management system for cloud computing, called VOLTAIC (Volume Optimization Layer To AssIgn Cloud resources). The proposal analyzes usage profiles of physical and virtual elements and defines heuristics based on differential utilization level that guarantee an optimized allocation of virtual elements. VOLTAIC introduces algorithms to determine proper parameters to allocate cloud elements and to automatically migrate those elements to avoid performance degradation due to server saturation. Results obtained through the implementation of the system in a small-scale cluster show that the system efficiently assigns virtual elements and ensures proper resource allocation to virtual elements. We also developed a virtual network simulator for cloud environments to attest the applicability of VOLTAIC in broader scenarios. Results show reductions in up to 10% in the amount of idle cycles due to correct assignment of virtual elements.