Test Platform Watch-Point Tagging for Message Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining correlation between input and output messages in distributed networks, such as those in the Internet of Things (IoT), are complex and require significant human effort or statistical data modeling, often resulting in inaccurate or intrusive code analysis tools that are not applicable for convenient and accurate correlation analysis.
Innovation Solution
A method and apparatus using watch-points and tags to determine correlation between input and output messages without performing semantic analysis or instrumenting codes, by setting watch-points for variables related to input messages and associating them with tags, allowing for correlation determination based on these tags when output messages are detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If semantic analysis is used to determine correlation between input and output messages, then correlation accuracy is improved, but device complexity and human effort increase significantly
Solution Approach 1:
The patent introduces tags as intermediary elements that link input and output messages without requiring semantic analysis. Each message is tagged with identifiers that enable correlation determination through simple tag matching, avoiding the need to analyze message contents semantically. This intermediary tagging mechanism resolves the contradiction by maintaining accuracy while eliminating complexity.
Solution Approach 2:
The patent creates simplified copies of message identifiers (tags) that represent the essential correlation information without containing the full semantic content. These tag copies enable fast correlation determination by comparing simple identifiers rather than analyzing complex message semantics, thus improving efficiency while maintaining correlation accuracy.
2Loss of information
If code analysis tools are used to find associated variables and functions, then correlation information is obtained, but the tools are either inaccurate, rigid, intrusive, or not generally applicable
Solution Approach 1:
The patent creates a universal tagging mechanism that can be applied across different programming languages, test platforms, and system types without modification. The tag-based approach is language-agnostic and platform-independent, enabling the same correlation determination method to work universally across diverse systems, thus resolving the adaptability issue while maintaining information completeness.
Solution Approach 2:
The patent extracts only the essential correlation information (message identifiers and tags) from the complex code execution environment, separating this key information from the intrusive code analysis process. By extracting and tagging messages at the I/O level rather than analyzing internal code structures, the method achieves correlation information completeness without the rigidity and intrusiveness of code-based tools.
3Reliability
If probes are embedded into executable machine codes at compiling time to record code dependency trees, then code execution paths are tracked, but the probes cannot be changed during code execution and the approach is rigid
Solution Approach 1:
The patent performs preliminary tagging of input and output messages with correlation identifiers before code execution begins. These tags are attached to messages at the I/O interface level prior to execution, enabling correlation tracking without requiring any probes or modifications during code execution. This preliminary action ensures reliable tracking while maintaining execution flexibility.
4Loss of time
If static code analysis is used to find associated variables and functions, then correlation information is obtained without executing codes, but the results are inaccurate for dynamic correlation determination
Solution Approach 1:
The patent introduces tags as intermediary markers that are attached to messages during actual code execution. These tags enable dynamic correlation determination by tracking which tagged input messages produce which tagged output messages during runtime, combining the speed of simple identification with the accuracy of dynamic execution-based correlation.
Data Source
AI summary
A method and apparatus for determining correlation between input and output messages in a system under test (SUT) is provided in the present invention. The SUT is provided with preset watch-points, and the running of the SUT is detected by triggering watch-points in a test platform at its run time. The method includes the steps of: upon detecting a message input operation, finding a variable that stores an input message, associating the variable with a tag of the input message, and adding a watch-point for the variable in the test platform; as well as, upon detecting network output operation, finding a variable that stores an output message of the SUT; and determining correlation between the output message and an input message according to a tag associated with the variable that stores the output message.