Cloud 3G Packet Core Control Plane with OpenFlow GTP Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3G packet core architectures face inefficiencies due to dedicated server pools, underutilization of resources, limited flexibility in peering points, and inability to support specialized data management for applications like deep packet inspection or transcoding, along with challenges in implementing GTP tunnel endpoint identifier routing using the OpenFlow protocol.
Innovation Solution
Implementing a cloud-based 3G packet core with a split architecture, where the control plane is executed in a cloud computing system using OpenFlow switches, enabling centralized management of the data plane and supporting GTP routing enhancements to facilitate multiple peering points, specialized service treatment, and efficient resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dedicated server pools are used for each operator network, then network reliability is improved, but resource utilization deteriorates and operational expenses increase
Solution Approach 1:
Multiple operator networks are merged into a single shared cloud facility, where control plane functions for different operators are consolidated on common hardware infrastructure. This merging eliminates the need for separate dedicated server pools for each operator, improving resource utilization while maintaining network reliability through virtualization and isolation mechanisms.
Solution Approach 2:
The cloud facility is designed with universal infrastructure that can serve multiple operator networks simultaneously. The control plane devices and virtual machines are configured to handle functions for different operators on the same physical hardware, making the system multi-functional and reducing overall resource consumption.
2Stability of the object's composition
If traditional 3G packet core architecture is used, then network stability is maintained, but flexibility in peering points and specialized data management is limited
Solution Approach 1:
The control plane is segmented into virtual machines that can be independently configured and deployed. This segmentation allows different virtual machines to handle different peering points and specialized data management functions simultaneously, providing flexibility while the underlying stable infrastructure maintains network reliability.
Solution Approach 2:
The system transitions from a static, fixed architecture to a dynamic cloud-based architecture where control plane functions can be dynamically allocated, scaled, and reconfigured. Virtual machines can be moved, created, or destroyed based on operational requirements, enabling flexible peering point implementation while maintaining stability through controlled orchestration.
3Device complexity
If GTP tunnel endpoint identifier routing is implemented using standard OpenFlow protocol, then protocol simplicity is maintained, but routing functionality is insufficient
Solution Approach 1:
A GTP extension module acts as an intermediary between the standard OpenFlow protocol and GTP routing requirements. This extension module translates and adapts OpenFlow messages to support GTP tunnel endpoint identifier routing, maintaining the simplicity of the base OpenFlow protocol while adding the necessary routing functionality through a layered approach.
Data Source
AI summary
A control plane device in a cloud computing system executes a plurality of virtual machines for implementing network function virtualization (NFV). The control plane device is operable to manage implementation of a general packet radio service (GPRS) tunnel protocol (GTP) in a packet core (PC) of a third generation (3G) network having a split architecture where a control plane of the PC of the 3G network is in the cloud computing system. The control plane communicates with a data plane of the PC through a control plane protocol. The data plane is implemented in a plurality of network devices of the 3G network. The control plane device and the plurality of virtual machines are operable to communicate with other control plane devices in the cloud computing system and with the plurality of network devices of the data plane.


