Application Framework Deployment Prediction for Incident Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Incident management in large multi-layer service-oriented platforms is challenging due to the complexity of interdependent services and microservices, making it difficult to understand potential causes and solutions for incidents triggered by code deployments.
Innovation Solution
An automated system that predicts incident mitigating changes by correlating data across various software platforms, using service dependency graphs and machine learning models to recommend code deployment modifications and identify potential incident-causing deployments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated prediction systems are implemented to analyze incident causes, then incident mitigation efficiency is improved, but system complexity and computational resource requirements increase
Solution Approach 1:
The system segments the complex incident analysis task into distinct functional modules: incident feature extraction module, data object identification module, candidate prediction generation module, and ranking module. Each module handles a specific aspect of the analysis, making the overall system more manageable and maintainable while improving incident mitigation efficiency
Solution Approach 2:
The patent introduces intermediate data structures (incident features, data objects, candidate predictions) that mediate between the raw incident data and the final mitigation recommendations. These intermediaries simplify the transformation process and make the system's internal workings more transparent and controllable
2Measurement precision
If comprehensive data correlation across multiple software platforms is performed, then prediction accuracy is improved, but data processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by pre-extracting incident features and pre-identifying relevant data objects from multiple software platforms before generating predictions. This preparation work is done in advance using machine learning models, reducing the computational burden during actual incident analysis and improving both accuracy and speed
Solution Approach 2:
The patent replaces manual mechanical analysis with automated machine learning models for feature extraction and data object identification. This substitution enables comprehensive data correlation across platforms while reducing processing time through intelligent algorithms that learn from historical incident patterns
3Device complexity
If manual investigation methods are used to identify incident causes, then system complexity remains low, but investigation time and resource consumption increase
Solution Approach 1:
The system enables self-service incident analysis by automatically generating prediction rankings and mitigation recommendations without requiring manual investigation. The machine learning models autonomously analyze incident data, identify relevant objects, and produce ranked predictions, freeing investigators from time-consuming manual analysis while maintaining manageable system complexity through modular design
Data Source
AI summary
Automated system for predicting software application framework deployment changes for incident mitigation is provided. An incident indication for an incident may be received. One or more incident features associated with the incident may be extracted. One or more data objects associated with the incident may be identified based on the one or more incident features. Candidate incident mitigation dataset comprising a plurality of incident mitigating predictions may be generated based on relation data and a topology graph structure associated with the one or more data objects. A ranked candidate incident mitigation dataset may be generated using one or more ranking models. One or more incident mitigating predictions from the ranked candidate incident mitigation dataset may be selected based on a rank value associated with each of the one or more incident mitigating predictions.