Load Testing Script Reference Detection for Dynamic Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current load testing methods require manual intervention to identify dynamic values, leading to a tedious and error-prone process that hinders efficiency and reliability.
Innovation Solution
A method and system utilizing a Large Language Model (LLM) to automatically determine references to dynamic values in load testing scripts by tokenizing and de-tokenizing natural language prompts, with user feedback for validation and LLM fine-tuning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual intervention is used to identify dynamic values in load testing scripts, then the process allows human control and validation, but it becomes tedious and error-prone, reducing efficiency and reliability
Solution Approach 1:
The system enables self-service by allowing the LLM to automatically identify dynamic values and generate references without requiring manual intervention. The load testing framework autonomously processes scripts, extracts dynamic values, and creates substitution references, eliminating the need for human operators to manually analyze and process testing scenarios.
Solution Approach 2:
The patent replaces the mechanical manual process of identifying dynamic values with an automated LLM-based system. The large language model processes natural language descriptions of testing scenarios and automatically generates accurate references to dynamic values, substituting human cognitive work with AI-based automated analysis.
2Productivity
If manual methods are used to determine references to dynamic values, then the process is simple in terms of technology, but it is tedious and error-prone, hindering efficiency
Solution Approach 1:
The LLM acts as an intermediary between the natural language testing scenario description and the load testing script execution. It translates human-readable scenario descriptions into structured references to dynamic values, bridging the gap between informal specifications and formal testing requirements without requiring complex manual parsing logic.
Solution Approach 2:
The system changes the parameter of processing speed by leveraging the LLM's ability to rapidly analyze and extract dynamic values from testing scenarios. This automated parameter change enables much faster identification of dynamic values compared to manual methods, while the complexity is managed through the LLM's inherent language understanding capabilities.
3Reliability
If automated LLM-based methods are used to identify dynamic values, then efficiency and accuracy are improved, but the system complexity increases due to tokenization and de-tokenization processes
Solution Approach 1:
The tokenizer performs preliminary action by converting the LLM's natural language output into a structured format that the load testing framework can process. This pre-processing step transforms the LLM's flexible language generation into deterministic references that can be reliably substituted in testing scripts, enabling automated processing while maintaining accuracy.
Solution Approach 2:
The de-tokenization process creates a copy of the structured reference data in a format suitable for substitution in load testing scripts. By copying and transforming the LLM's output through tokenization and de-tokenization, the system maintains the accuracy benefits of AI-based identification while adapting the data to the specific requirements of the testing framework.
4Loss of time
If extensive manual processing is used to handle dynamic values, then detailed control is possible, but it leads to time-consuming and labor-intensive processes
Solution Approach 1:
The system performs self-service by automatically completing the entire process of identifying dynamic values, generating references, and preparing load testing scripts without requiring manual intervention. The LLM autonomously processes testing scenarios and generates executable testing configurations, eliminating time-consuming manual operations while maintaining ease of use through natural language interaction.
Solution Approach 2:
The automated LLM-based system enables continuous processing of load testing scenarios without the interruptions and transitions required by manual methods. The system can continuously receive testing scenario descriptions, process them through the LLM, generate references to dynamic values, and prepare testing scripts in an uninterrupted automated workflow, significantly reducing setup time.
Data Source
AI summary
A method and system of determining reference to dynamic values is disclosed. A processor receives at least one testing scenario for load testing an application. At least one load testing script is determined based on the at least one load testing scenario. A natural language prompt is received from a user for identifying the reference to the at least one dynamic value within the at least one load testing script. The at least one load testing script and the natural language prompt is tokenized to generate a tokenized load testing script and a tokenized prompt. An LLM is prompted using the tokenized prompt to determine at least one tokenized reference corresponding to at least one tokenized dynamic value from the tokenized load testing script. The at least one tokenized reference is de-tokenized to determine an LLM determined reference to the at least one dynamic value.