Controller-to-Controller Interface for Layer-Mapped Topology Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-layer networks face challenges in efficiently communicating and sharing information between network layers due to the lack of layer-specific optimizations in common representations, which hinders independent evolution and faster processing.
Innovation Solution
Implementing a controller-to-controller interface (CCI) and layer mapping function (LMF) to translate network topology maps between layers, using a common link abstraction model (CAM) that maintains layer-specific optimizations, allowing for efficient communication and independent evolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single common control-plane representation (GMPLS) is used for multiple network layers, then interoperability between layers is improved, but layer-specific optimizations and independent evolution are lost
Solution Approach 1:
The patent segments the network control plane into multiple independent controllers, each responsible for a specific network layer. Each controller maintains its own layer-specific data structures and protocols, allowing independent optimization and evolution while still enabling cross-layer communication through defined interfaces.
Solution Approach 2:
The patent introduces an intermediary mechanism (controller interface and message exchange protocol) that enables communication between layer-specific controllers without requiring a unified common representation. This intermediary layer translates and adapts messages between different layer protocols, preserving both interoperability and layer-specific characteristics.
2Productivity
If layer-specific protocols and formats are used for each network layer, then processing speed and layer optimization are improved, but communication and information sharing between layers become difficult
Solution Approach 1:
The patent segments the control plane into independent layer-specific controllers that process information at optimal speeds using layer-appropriate protocols. Each controller maintains full awareness of its layer's state while independently processing information, eliminating the need for continuous translation and interpretation that would slow down processing.
Solution Approach 2:
The patent establishes feedback mechanisms where each layer controller reports its state and receives requests from other layers through standardized interfaces. This feedback loop ensures that information is shared between layers while maintaining the efficiency of layer-specific processing, as each controller can respond to cross-layer requests using its own optimized protocols.
Data Source
AI summary
A controller at an IP (e.g., client) layer in a multi-layer network can request a network topology map from another controller at an optical (e.g., server) layer in the multi-layer network. The controller at the optical layer of the network can use a layer mapping function and common attributes between the formats used to describe the network topology map at the two layers to generate a common layer abstraction model representing the network topology map stored at the controller at the optical layer of the network. A controller-to-controller interface can translate and/or send the common layer abstraction model to the controller at the IP layer for processing data on the network.