Network Function Chain Routing via Graph Cost Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network function chaining methods lack flexibility and efficiency in routing traffic flows through dynamic network function chains, especially with the introduction of network function virtualization (NFV), which complicates the determination of optimal paths due to varying costs and states of network functions across multiple instances.
Innovation Solution
A method involving graph manipulation, where subgraphs representing network elements are connected to form a graph with cost measures, allowing for the selection of a path through network functions based on cost optimization, considering both deployed and planned instances, and accommodating flexible ordering of network functions in the chain.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If network function virtualization is implemented to allow dynamic deployment of network functions, then flexibility and adaptability of network function chaining is improved, but device complexity and difficulty of determining optimal paths increases
Solution Approach 1:
The patent segments the network function chaining problem into graph construction and path selection components. The graph is built by representing network elements as vertices and connections as edges, with each network function instance becoming a node in the graph. This segmentation transforms the complex routing determination into a structured graph problem that can be solved using standard pathfinding algorithms.
Solution Approach 2:
The patent introduces a graph as an intermediary data structure between the network function deployment and the routing decision. The graph serves as a mediator that abstracts the complex network state into a simplified representation where vertices represent network elements and edges represent possible routing paths, allowing optimal paths to be determined through cost-based evaluation.
2Productivity
If multiple instances of network functions are deployed at various locations for load balancing and latency reduction, then network performance is improved, but the complexity of evaluating network states and determining optimal routes increases
Solution Approach 1:
The patent segments the evaluation of multiple network function instances by creating separate graph representations for different deployment scenarios. Each graph captures the state of network elements and their connections, allowing the system to evaluate multiple instances independently and then select the optimal combination through pathfinding algorithms that consider cost measures.
Solution Approach 2:
The patent changes the parameter representation by introducing cost measures associated with graph edges and vertices. These cost parameters encode information about network state, performance metrics, and routing preferences, transforming the complex evaluation of multiple instances into a parameter-based optimization problem that can be solved efficiently.
3Productivity
If dynamic routing based on cost measures is implemented, then routing efficiency and cost optimization are improved, but the complexity of graph manipulation and path selection increases
Solution Approach 1:
The patent implements feedback through cost measures that are continuously evaluated based on network state. The graph structure allows routing decisions to be made based on current cost information, and the system can re-evaluate and adjust paths as network conditions change, providing feedback-driven optimization without requiring complex real-time graph manipulation.
Solution Approach 2:
The patent performs preliminary graph construction and cost evaluation before actual routing decisions are made. By pre-processing the network state into a graph structure with associated cost measures, the system simplifies the subsequent path selection process, avoiding the need for complex graph manipulation during time-critical routing operations.
Data Source
AI summary
In one embodiment, a request is received to route a network function chain. For each contained network function, a subgraph is generated, where each of a plurality of network elements in a network is split into two vertexes, and one edge is added between the split two vertexes for each network element that hosts that network function of the subgraph. The subgraphs are ordered and connected through connecting each vertex with one edge to another vertex with one edge in a subsequent subgraph to form a graph, where the connection is included in a representation of the network. Each edge includes a cost measure. The method selects a path from the vertex representing the source network element to the vertex representing the destination network element in the graph to route the network function chain, where each edge of the path is selected based on at least its cost measure.