Dynamic Graph-Based Network Management for NFV and SDN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing complexity of telecommunications network infrastructures due to emerging technologies like NFV and SDN poses challenges for network service providers in managing and maintaining high-quality services, as conventional OSSs struggle to adapt to diverse network architectures and require costly updates and complex collaborations.
Innovation Solution
A graph-based data structure system that represents network entities and relationships dynamically, using a graph processing engine to construct and update network graphs based on provider-specific data, allowing for incremental processing and efficient management of network entities and connections, and enabling dynamic execution of walks to generate output entities for application services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional OSSs are used to manage network infrastructures, then existing network management functionality is maintained, but the system cannot adapt to emerging technologies like NFV and SDN, requiring costly updates and complex collaborations
Solution Approach 1:
The patent implements a dynamic graph-based data structure that automatically adapts to emerging network technologies like NFV and SDN. The graph structure allows flexible representation of virtual network functions, physical network elements, and their relationships, enabling the OSS to evolve with network architectures without requiring complex reconfiguration or costly updates.
Solution Approach 2:
The graph-based framework provides a universal model that can represent multiple network technologies and architectures within a single unified structure. This multi-functional approach allows the same OSS platform to manage traditional network elements alongside virtualized functions, eliminating the need for separate management systems and reducing operational complexity.
2Productivity
If network infrastructure is expanded to accommodate new technologies, then network capacity and service offerings are increased, but operational complexity increases
Solution Approach 1:
The patent segments network management into modular graph-based representations where each network element, virtual function, or connection is represented as distinct graph nodes and edges. This segmentation allows operators to manage complex networks by working with discrete, well-defined units and their relationships, reducing operational complexity while supporting expanded network capacity.
Solution Approach 2:
The graph-based data structure serves as an intermediary layer between network operations and the underlying complex infrastructure. It abstracts the complexity of NFV, SDN, and traditional networking into a unified representation, enabling operators to interact with the network through simplified graph queries and operations rather than dealing with direct infrastructure complexity.
3Ease of manufacture
If graph-based data structure is implemented, then network management flexibility and efficiency are improved, but system complexity increases
Solution Approach 1:
The patent uses a simplified graph-based representation that copies essential network management functionality into a standardized data structure format. This approach captures the necessary management capabilities in a universal graph model, making network management easier while avoiding the complexity of implementing multiple specialized systems for different network technologies.
Data Source
AI summary
Embodiments are disclosed for providing network management and application services for a telecommunications network. In an embodiment, data associated with a network entity in the telecommunication network may be received from a provider-specific data source, and the received data may include one or more data attribute values. The telecommunications network may be represented by a network graph containing vertices and edges, and each vertex may correspond to a respective network entity having a respective entity type. The plurality of edges in the network graph connected to the first vertex may be updated based on a linking rule that specifies a relationship between network entities having respective entity types, and a walk associated with the first vertex may be identified. The identified walk may then be dynamically executed from an intermediate starting vertex to generate one or more output entities.