Service Function Chain Dynamic Classification via Predictive Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In service function chaining environments, conventional classifiers may fail to apply appropriate service functions to additional network traffic triggered by original traffic, leading to undesirable results such as dropped packets or inadequate security, as they lack pre-programmed rules for dynamic traffic patterns.
Innovation Solution
The implementation of dynamic classification instruction modules in service nodes that predict additional network traffic and generate instructions for dynamically assigning a service function path to the classifier, ensuring that both original and predicted traffic follow the same or appropriate service function paths, using Network Service Headers (NSH) for communication between service nodes and the classifier.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional classifiers with pre-programmed rules are used, then classification is simple and fast, but additional network traffic triggered by original traffic cannot be properly classified leading to packet drops or inadequate security
Solution Approach 1:
The system performs preliminary classification of original network traffic to determine the service function path, then uses this information to proactively predict and classify additional triggered traffic before it reaches the classifier, ensuring both traffic types receive appropriate service function processing
Solution Approach 2:
The system implements a feedback mechanism where the classifier communicates with service nodes about classified traffic patterns, and service nodes provide information about additional triggered traffic, allowing the classifier to dynamically adjust its classification rules and improve future classification accuracy
2Reliability
If dynamic classification rules are implemented to handle additional traffic, then traffic classification accuracy improves, but system complexity increases
Solution Approach 1:
The system introduces service nodes as intermediaries between the original traffic source and the classifier. These service nodes monitor traffic patterns, identify additional triggered traffic, and communicate with the classifier to enable dynamic classification without requiring the classifier itself to become more complex
Solution Approach 2:
The service nodes perform multiple functions: they process original traffic through service functions, monitor for additional triggered traffic, and communicate classification information to the classifier. This multi-functionality reduces the need for separate specialized components, managing system complexity while improving classification accuracy
Data Source
AI summary
In one example, a service function forwarder of a service function chain enabled domain receives, from a classifier of the service function chain enabled domain, network traffic assigned to a service function path that includes at least one service node configured to apply a service function to the network traffic. The service function forwarder forwards the network traffic along the service function path. The service function forwarder receives, from the at least one service node, instructions for dynamically assigning a particular service function path to predicted network traffic that the at least one service node predicts will be triggered by the network traffic. The service function forwarder forwards the instructions to the classifier.