No abstract

No abstract

The IP Multimedia System (IMS) is an important reference service delivery platform for next generation networks and is considered as a de-facto standard for IP-based multimedia communication services. In its current design, the IMS faces important challenges in terms of scalability and elasticity, and lacks the ability to adaptively manage the network resources and dynamically dimension the network nodes based on load and demand. Network function virtualization and cloud computing are two important concepts that can be leveraged to address those challenges in IMS environments. In this work, we propose two adaptive SLA-based elasticity management algorithms for virtualized IMS environments. Our proposed algorithms use two SLA attributes (the call setup delay and user priority) to dynamically control the CPU resources allocated/de-allocated to virtualized IMS nodes. The aims of our proposed algorithms are: 1) to ensure efficient usage and sharing of CPU resources by various IMS components; 2) to reduce the overall power consumption in virtualized IMS platforms; and 3) to enhance the user experience when using IMS networks. We have tested the proposed algorithms by setting up a virtualized IMS environment using OpenIMS Core and Xen as the hypervisor. The results obtained show that our proposed algorithms meet the SLA constraints, even when subjected to dynamic load, thereby enhancing the overall QoS. We have also compared the proposed algorithms with Xen Server's existing CPU resource scaling governors and the results indicate that our algorithms work better when compared to the existing governors.

No abstract