Application-Agnostic Fault Detection via Statistical Event Interception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fault detection methods require extensive customization, are limited to specific fault assumptions or models, and fail to detect 'soft faults' where applications appear functional but are impaired, lacking transparency and generality across systems and fault scenarios.
Innovation Solution
A statistically based, application-agnostic fault detection system that builds a dynamic model of running applications without pre-defined knowledge, operating transparently and automatically, capable of detecting faults in both the application and its environment, using interception layers and statistical fault detectors to compare run-time characteristics against historical norms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fault detection methods are used, then specific fault scenarios can be detected, but extensive customization and application-specific modifications are required
Solution Approach 1:
The patent implements a universal fault detection framework that monitors multiple application types (database, web server, file server, etc.) using a single standardized architecture. The system collects performance metrics, generates events, and detects faults across diverse applications without requiring application-specific customization, thereby achieving broad universality while maintaining detection accuracy.
Solution Approach 2:
The patent introduces an intermediary event processing layer that sits between raw performance metrics and fault detection logic. This event processing module standardizes the interface between applications and the fault detection system, converting diverse application metrics into uniform events that can be processed by generic fault detection rules, eliminating the need for application-specific modifications.
2Reliability
If application-specific fault detectors are implemented, then detection accuracy for known faults improves, but the system cannot detect soft faults or faults in unknown applications
Solution Approach 1:
The patent enables the fault detection system to automatically adapt to new application types without manual configuration. The system self-services by collecting performance metrics from any application, automatically generating appropriate events, and applying generic fault detection rules. This self-adapting capability allows the system to detect both known and unknown faults across diverse applications, achieving high reliability and versatility simultaneously.
Solution Approach 2:
The patent implements dynamic fault detection that adapts to changing application behaviors and environments. The system continuously monitors performance metrics, dynamically generates events based on observed patterns, and adjusts detection thresholds based on historical data. This dynamic approach enables the system to detect soft faults and adapt to new applications, maintaining high reliability across varying conditions.
3Measurement precision
If manual fault detection configuration is used, then detection rules can be precisely tailored, but the system requires significant setup time and expert knowledge
Solution Approach 1:
The patent pre-configures a comprehensive set of generic fault detection rules and event processing logic that can be immediately applied to any application. The system includes pre-defined performance thresholds, event generation templates, and detection algorithms that are ready to use upon deployment. This preliminary preparation eliminates the need for time-consuming manual configuration while maintaining detection precision through the standardized rules.
Solution Approach 2:
The patent enables automatic rule generation and configuration based on observed application behavior. The system automatically learns performance patterns from collected metrics, generates appropriate detection rules, and configures itself without requiring expert intervention. This self-service capability significantly reduces deployment time while maintaining precise detection through data-driven rule generation.
4Difficulty of detecting and measuring
If the fault detection system modifies the application, then detection capability improves, but the application requires customization and loses transparency
Solution Approach 1:
The patent introduces an intermediary event processing layer that collects performance metrics from applications without modifying their core logic. The system uses standard interfaces and APIs to gather metrics, converts them into events, and processes them through generic fault detection rules. This intermediary approach enables comprehensive fault detection capability while maintaining application transparency, as applications continue to operate unchanged.
Data Source
AI summary
A system, method, and computer readable medium for statistical application-agnostic fault detection of multi-process applications. The computer readable medium includes computer-executable instructions for execution by a processing system. A multi-process application runs on a host. Interceptors collect statistical events and sends said events to a statistical fault detector. The statistical fault detector creates one or more distributions and compares recent statistical event data to historical statistical event data and uses deviation from historical norm for fault detection. The present invention detects faults both within the application and within the environment wherein the application executes, if conditions within the environment cause impaired application performance. The invention also teaches consensus fault detection and elimination of cascading fault notifications based on a hierarchy of events and event groups. Interception and fault detection is transparent to the application, operating system, networking stack and libraries.