Network Verification Using Symbolic Packets and Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional network verification methods are resource-intensive and impractical for large-scale cloud computing networks, as they require replicating the entire network and analyzing extensive network characteristics, which is time-consuming and costly, and lack versatility in testing network changes effectively.
Innovation Solution
A method and system for network verification that uses a general-purpose programming language to develop tests, allowing for the creation of symbolic packets and tailored network models focused on specific changes, reducing processing resources and increasing test efficiency by only analyzing relevant network elements and data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional network verification replicates the entire network to test changes, then verification completeness is improved, but resource consumption and time cost increase significantly
Solution Approach 1:
The patent segments the network into relevant and irrelevant portions relative to the change being tested. Instead of replicating the entire network, the system identifies and isolates only the network elements that are directly affected by the proposed change, creating a minimized test network that maintains verification effectiveness while reducing resource consumption.
Solution Approach 2:
The patent extracts only the necessary network elements required to test the specific change from the complete network. By removing irrelevant network portions and focusing solely on the affected elements, the system achieves verification completeness for the specific change without the overhead of replicating the entire network.
2Reliability
If conventional network verification replicates the entire network, then all network pathways are tested, but processing resources and complexity increase
Solution Approach 1:
The verification system is segmented to operate on different levels: a high-level network model for planning and a minimized test network for actual verification. This segmentation allows the system to maintain comprehensive verification coverage while reducing the complexity of the test execution environment.
Solution Approach 2:
The patent creates a simplified copy or model of only the relevant network portions rather than replicating the entire network. This selective copying reduces the complexity of the verification system while maintaining the ability to test all necessary pathways affected by the change.
3Measurement precision
If static analysis is used to analyze network packets hop by hop, then network transformation is observed, but extensive preprocessing is required
Solution Approach 1:
The patent performs preliminary actions by pre-identifying relevant network elements and constructing the minimized test network model before actual verification begins. This preliminary setup phase reduces the preprocessing burden during actual verification execution, as the system only needs to test within the already-filtered relevant portions.
Solution Approach 2:
The patent changes the parameters of network element representation from detailed implementation specifics to abstracted relevant properties. By transforming network elements into a standardized format that captures only verification-relevant characteristics, the system reduces preprocessing complexity while maintaining measurement precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A network verification system uses general-purpose programming language to create network verification tests. A test orchestrator builds a model of the network only using data from the network verification test. An optimization testing manager creates symbolic packets for verification tests using assertions based on a packet library embedded into the testing manager and the general-purpose programming language.