SDN Control Plane Policy Generation for Traffic Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network technologies face limitations in efficiently managing and optimizing network traffic within software-defined networks (SDNs), particularly in ensuring performance, reliability, and security, due to the separation of control and data planes, which can lead to suboptimal routing and resource utilization.
Innovation Solution
A method that involves monitoring data from SDN devices, generating data models based on SDN parameters, and creating policies to steer traffic effectively across the network, using the control plane to isolate and manage data and administrative traffic, thereby optimizing path selection and resource allocation based on performance, cost, and security considerations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the control plane and data plane are separated in SDN, then network flexibility and programmability are improved, but network complexity increases
Solution Approach 1:
The patent segments the network into control plane and data plane, with the control plane handling policy decisions and the data plane handling traffic forwarding. This separation allows independent optimization of each plane while maintaining overall network flexibility.
Solution Approach 2:
The patent introduces a control device as an intermediary between the data plane devices and the external control system. This mediator simplifies the complexity by providing a centralized point for policy management and coordination, reducing the burden on individual data plane devices.
2Reliability
If application-aware routing is implemented, then user experience is improved, but processing requirements increase
Solution Approach 1:
The patent implements preliminary classification of traffic flows at the data plane devices based on simple criteria such as source/destination addresses and port numbers. This preliminary action reduces the processing burden by pre-categorizing traffic before it reaches the application-aware routing logic in the control plane.
Solution Approach 2:
The patent replaces complex application-layer inspection with lower-layer packet classification mechanisms. By using network layer and transport layer information (IP addresses, ports, protocols) instead of deep packet inspection, the system achieves application-aware routing with reduced processing requirements.
3Loss of information
If monitor data is collected from all devices, then network visibility is improved, but communication overhead increases
Solution Approach 1:
The patent implements selective monitoring where different devices report different types of monitor data based on their local characteristics and relevance to current network conditions. Not all devices report all types of data at all times, reducing overall communication overhead while maintaining necessary network visibility.
Solution Approach 2:
The patent implements periodic collection of monitor data rather than continuous reporting. Devices report their status at scheduled intervals or when significant changes occur, reducing communication overhead while maintaining adequate network visibility for control decisions.
Data Source
AI summary
A method may include receiving monitor data via the control plane from at least one device of a set of devices in the SDN. The method may further include generating a data model based on a set of SDN parameters and the monitor data. The method may also include determining a change for at least one device of the set of devices in the SDN based on the data model. The method may include generating a policy, based on the change for at least one device of the set of devices in the SDN. The method may further include sending the policy via the control plane to the set of devices in the SDN.