Query Metric Scoring for Faster Database Performance Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying the cause of poor query performance in database systems is difficult and time-consuming, affecting database efficiency and user experience.
Innovation Solution
A system and method for scoring queries using sub-scores for metrics like latency, scan, and payload, generating an overall score, and providing a user interface to display and customize query performance data, enabling efficient identification and improvement of query performance issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional query performance monitoring methods are used, then database efficiency is maintained, but the time required to identify performance issues increases significantly
Solution Approach 1:
The patent segments query performance into multiple discrete metrics (latency, scan count, rows returned, CPU time, I/O operations) and evaluates each metric independently with weighted scoring. This segmentation allows rapid identification of specific performance bottlenecks without requiring complex holistic analysis, reducing the time to diagnose issues while maintaining systematic monitoring capability.
Solution Approach 2:
The system implements continuous feedback loops where query performance metrics are collected, analyzed, and fed back to users through dashboards and alerts. This automated feedback mechanism enables real-time identification of performance degradation without manual intervention, significantly reducing detection time while the structured feedback framework prevents system complexity from becoming unmanageable.
2Measurement precision
If detailed query metrics are collected and analyzed, then measurement precision of performance issues is improved, but the complexity of the monitoring system increases
Solution Approach 1:
The patent divides query performance measurement into distinct, well-defined metric segments (latency, scan count, rows returned, CPU time, I/O operations). Each segment is measured independently with clear definitions and calculation methods, achieving high measurement precision without requiring a monolithic complex system. The segmented approach allows selective collection and analysis of specific metrics based on needs.
Solution Approach 2:
The system employs parameter-based scoring where each metric is normalized and weighted according to its importance. By transforming raw metric values into standardized scores with configurable weights, the system achieves precise performance characterization while maintaining flexibility. This parameter transformation approach simplifies the monitoring system architecture compared to raw metric analysis.
3Productivity
If comprehensive query performance data is provided to users, then the ability to resolve performance issues is improved, but the amount of information to be processed increases
Solution Approach 1:
The patent segments comprehensive performance data into organized metric categories with hierarchical structure. Users can access detailed information when needed while the segmented presentation allows quick scanning of key performance indicators. This segmentation enables efficient data processing by users without overwhelming them with undifferentiated information volumes.
Solution Approach 2:
The system applies local quality by providing different levels of detail in different contexts. Summary views present only critical metrics for quick assessment, while detailed analysis is available on-demand for specific metrics. This localized information quality allows users to process data efficiently at the appropriate level of detail for their current needs, improving resolution speed without requiring all users to process complete data sets.
Data Source
AI summary
The present disclosure relates generally to scoring queries based on accesses to databases. The scoring of a query can include generating a sub-score for each of one or more query metrics and generating an overall score for the query using at least some or all of the sub-scores. Additional information that is associated with the query may also be determined. User inputs can be received to modify the scoring of the queries, to modify the information presented in a user interface, and/or to modify the determination of some or all of the additional information.