Synthetic KPI Metric Test Data Generation for Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing systems require a significant amount of time to generate key performance indicator (KPI) metric data, which delays validation assessments such as performance tests, as they need to operate for a period to collect actual measurements, thereby increasing the time required for system managers to conduct tests.
Innovation Solution
A method and system for generating KPI metric test data that allows for prompt testing of interface layers or service logic without the need for actual data collection time, by receiving KPI parameter data and using it to generate test data values based on a defined time period and collection frequency, enabling complete control over the generation of KPI metric data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the computing system operates for a period of time to collect actual KPI measurements, then realistic KPI measurement data is generated, but the time required for validation assessments increases
Solution Approach 1:
The patent creates synthetic KPI test data that copies the structure, format, and statistical characteristics of real KPI measurements without requiring actual system operation. The generated data mimics real measurement patterns including time series correlations, value distributions, and missing data patterns, enabling validation assessments to proceed immediately with realistic test data.
Solution Approach 2:
The system pre-generates KPI test data before validation assessments are needed. By creating realistic KPI measurement data in advance through synthetic generation methods, the system eliminates the waiting period that would otherwise be required to collect actual measurements, allowing immediate execution of validation assessments.
2Loss of time
If the computing system is configured to collect data over a short period of time and subsequently copied to represent a larger time period, then the time required for data collection is reduced, but the realism and accuracy of KPI measurement data deteriorates
Solution Approach 1:
The patent transforms the approach by changing from copying actual short-term data to generating synthetic data with controlled statistical parameters. The system uses parameterized models that incorporate domain knowledge about KPI behavior, time series characteristics, and system-specific patterns to generate realistic test data without being constrained by actual collection periods.
Solution Approach 2:
The synthetic data generation applies different quality characteristics to different aspects of the generated data. The system maintains high realism in statistical distributions, temporal correlations, and value ranges while allowing flexibility in the overall time period coverage, enabling realistic test data generation independent of actual system operation duration.
Data Source
AI summary
Methods, systems, and computer readable mediums for generating key performance indicator (KPI) metric test data are disclosed. According to one method, the method includes receiving KPI parameter data associated with a KPI metric corresponding to a system element hosted by a computing system, generating KPI metric test data values for the KPI metric in accordance to a defined time period and a collection frequency interval specified by the KPI parameter data, and utilizing the KPI metric test data values to subject at least one of an interface layer or service logic of the computing system to a validation assessment.