Window Deviation Analyzer for Performance Regression Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Developers lack tools to identify and rectify performance regressions in distributed systems during the development cycle, making it difficult and costly to track the cause of performance issues after integrating offending code.

Innovation Solution

A method and system that analyze data samples within a historic and recent time window to detect performance regressions by calculating mean or median deltas and transmitting alerts when these exceed user-defined tolerance thresholds, allowing for real-time notification of performance issues.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If developers manually track performance regressions after code integration, then they can identify performance issues, but the process becomes difficult, costly, and time-consuming

Engineering Contradiction:

Improveperformance regression detection reliabilityVSAvoidtime to identify and track performance regression

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by establishing baseline performance metrics before code integration and continuously monitoring performance data. This allows the system to detect regressions automatically without requiring developers to manually track performance issues after integration, thereby reducing time loss while maintaining detection reliability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by comparing current performance metrics against historical baselines and automatically notifying developers of regressions. This automated feedback loop eliminates the need for manual tracking and significantly reduces the time required to identify performance issues while maintaining high detection reliability.

Inventive Principle:

Principle #23Feedback

2Difficulty of detecting and measuring

If developers work backward to track performance regression causes, then they can identify the source of issues, but the complexity and cost increase significantly

Engineering Contradiction:

Improvedifficulty of tracking performance regression causeVSAvoidcomplexity of performance tracking system

Core Design Contradiction:

Difficulty of detecting and measuringVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by implementing automated tracking and logging of performance metrics with timestamps and code version associations. This preliminary setup enables forward-looking detection of regression sources rather than requiring backward tracing, reducing detection difficulty while managing system complexity through structured data collection.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediary automated analysis layer that processes performance data, correlates it with code changes, and identifies regression sources. This intermediary component simplifies the detection process by handling the complex analysis work automatically, reducing both detection difficulty and the apparent complexity for developers.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Loss of time

If performance monitoring is implemented continuously, then performance regressions can be detected early, but data processing requirements and system overhead increase

Engineering Contradiction:

Improvetime to detect performance regressionVSAvoidcomputational resources for data processing

Core Design Contradiction:

Loss of timeVSUse of energy by moving object

Solution Approach 1:

The system applies partial action by monitoring only critical performance metrics and using selective analysis rather than processing all possible data. This approach enables early detection of performance regressions while minimizing computational overhead by focusing resources on the most important performance indicators.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system dynamically adjusts monitoring parameters and thresholds based on system state and historical data. This allows the system to maintain sensitive detection capabilities when needed while reducing processing intensity during normal operation, thereby balancing early detection capability with computational resource consumption.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10157116B2Window deviation analyzer

Publication Date: 2018.12.18 GOOGLE LLC

AI summary

A method includes receiving samples of data over a time range from a start time to an end time, determining a mean/median of the data samples within a historic time window, and determining a mean/median of the data samples within a recent time window. The method also includes determining a data sample delta based on a difference between the mean/median of the data samples within the recent time window and the mean/median of the data samples within the historic time window. The method also includes determining whether the data sample delta exceeds a delta tolerance, and when the data sample delta exceeds the delta tolerance, determining a regression in performance. The method also includes transmitting a regression alert to a user device. The user device is configured to notify a user associated with the user device of the regression in performance.