Fault Localization in Cloud-Native Applications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current anomaly detection and fault localization techniques for cloud-native applications are inefficient in identifying potential problems and narrowing down causes of issues in a timely and accurate manner, particularly in cloud-native and hybrid environments.
Innovation Solution
The method involves classifying event-related alerts, processing application logs to identify error logs, and using topological sorting of run-time service invocation graphs to isolate faults, while analyzing log data to distinguish between error-related and normal behaviors, and inferring faulty components by extracting entities from log messages and analyzing request flows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional anomaly detection and fault localization techniques are used for cloud-native applications, then the system can detect potential problems, but the identification of potential problems and narrowing down causes is inefficient and not timely
Solution Approach 1:
The patent segments the fault localization process into distinct phases: alert classification, log retrieval, log processing, error identification, and prioritization. Each phase handles specific tasks independently, improving overall efficiency. The system segments logs by type (application logs, system logs, network logs) and processes them through specialized handlers, enabling parallel processing and reducing time to identify fault causes.
Solution Approach 2:
The patent introduces an intermediary fault localization system that sits between anomaly detection and root cause analysis. This intermediary classifies alerts, retrieves relevant logs, processes them through multiple handlers, and prioritizes errors before presenting to operators. This intermediary layer automates the time-consuming manual processes of log gathering and analysis, significantly reducing the time to narrow down fault causes while maintaining high productivity.
2Measurement precision
If comprehensive log analysis is performed to accurately identify faulty components, then diagnostic accuracy improves, but the complexity of processing multiple application logs increases
Solution Approach 1:
The patent divides the log processing system into specialized handlers for different log types (application logs, system logs, network logs). Each handler is responsible for specific processing tasks appropriate to its log type, reducing the complexity of any single handler while maintaining comprehensive analysis capability. This segmentation allows the system to achieve high diagnostic accuracy without overwhelming complexity in each individual processing component.
Solution Approach 2:
The fault localization system acts as an intermediary that manages the complexity of comprehensive log analysis. It automatically retrieves, filters, and processes multiple application logs through appropriate handlers, then synthesizes results into prioritized error lists. This intermediary approach maintains high measurement precision for fault identification while abstracting away the processing complexity from end users and simplifying the overall system architecture.
3Productivity
If manual analysis of event-related alerts and logs is performed, then detailed fault diagnosis can be achieved, but the process is time-consuming and less efficient
Solution Approach 1:
The patent implements self-service automation where the fault localization system automatically performs alert classification, log retrieval, log processing, error identification, and prioritization without manual intervention. The system serves itself by autonomously navigating the complex processes of multi-log analysis and fault identification, dramatically improving diagnostic efficiency and reducing the time required for fault diagnosis while maintaining detailed analysis capabilities.
Data Source
AI summary
Methods, systems, and computer program products for fault localization for cloud-native applications are provided herein. A computer-implemented method includes classifying an event-related alert directed to at least one system by processing one or more characteristics of the event-related alert; obtaining and processing multiple application logs based at least in part on the classification of the event-related alert; identifying error logs among the multiple application logs based at least in part on the processing of the multiple application logs; ordering the error logs using one or more prioritization techniques; and performing at least one automated action based at least in part on the ordering of the error logs.