Scalable Evolved Packet Core with Distributed Virtualized Planes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional evolved packet core (EPC) network deployments do not allow for independent scaling of data and control aspects, leading to inefficient resource utilization and increased costs when handling varying bandwidth demands, such as those from 4K video applications.
Innovation Solution
A scalable EPC solution is implemented using a distributed core framework with separate virtualized control and data planes, allowing for elastic and dynamic scaling of data and signaling aspects independently, using virtual machines and shared network interfaces to manage sessions and optimize resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If dedicated physical devices are used to handle both signaling and data aspects, then device functionality is complete, but resource utilization efficiency deteriorates when only data aspect needs to be increased
Solution Approach 1:
The patent segments the EPC device into two independent virtual machines: a control plane VM handling signaling aspects and a data plane VM handling data aspects. This segmentation allows each plane to be scaled independently based on specific needs, resolving the contradiction by enabling adaptive scaling without forcing unnecessary increases in both planes simultaneously.
Solution Approach 2:
The patent creates virtual machine instances that copy and virtualize the traditional physical EPC device functions. By using virtualized copies rather than dedicated physical devices, the system achieves flexible resource allocation and independent scaling of control and data planes, improving resource utilization while maintaining complete functionality.
2Productivity
If a new device is added to increase data aspect, then data handling capacity is improved, but device complexity and configuration difficulty increase due to integrated control and data aspects
Solution Approach 1:
By segmenting the EPC into separate control plane and data plane virtual machines, the patent eliminates the need to configure integrated devices. When data handling capacity needs to be increased, only the data plane VM needs to be scaled or added, without affecting control plane configuration, thereby reducing device complexity and configuration difficulty.
Solution Approach 2:
The patent extracts the data plane functionality from the integrated control and data plane device. This extraction allows the data plane to be independently scaled or added without bringing in unnecessary control plane complexity, simplifying the overall system configuration while improving data handling capacity.
3Reliability
If dedicated physical devices are used, then signaling and data handling are reliable, but scalability deteriorates when only one aspect needs to be increased
Solution Approach 1:
The patent uses virtual machine instances that replicate the functionality of dedicated physical devices while adding virtualization flexibility. The virtualized control and data planes maintain reliable signaling and data handling through proper virtualization implementation, while simultaneously enabling elastic scaling by independently deploying or scaling data plane VMs without affecting control plane reliability.
Solution Approach 2:
The patent transforms the static, fixed configuration of dedicated physical devices into a dynamic virtualized environment. The control and data planes can be dynamically scaled, deployed, and configured as virtual machine instances, maintaining reliability through virtualization while enabling elastic adaptation to changing network demands.
Data Source
AI summary
The techniques described herein relate to methods, apparatus, and computer readable media configured to provide a distributed core framework for a voice and data network. A control plane comprising a set of control plane components is executed using a set of virtual machines running on a set of computing devices. The control plane comprises a first network interface to the voice and data network that is shared by the set of control plane components. A data plane comprising a set of data plane components is executed using a set of virtual machines running on a set of computing devices. The data plane comprises a second network interface to the voice and data network that is shared by the set of data plane components. Upon receipt of a session request from a remote device, a selected data plane component is selected to handle a corresponding session, such that the selected data plane component can directly communicate with the remote device using the second network interface to handle the session.