AI Cloud Infrastructure Component Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud environment resolution techniques lack a systematic approach for prioritizing multi-tenant incidents, often relying on human estimation and resulting in cost-intensive delays and outages due to arbitrary selection of resolution order.
Innovation Solution
The implementation of artificial intelligence techniques to automatically classify cloud infrastructure components, using historical data to prioritize incident-related resolution by training machine learning algorithms and neural networks to predict the priority class of runtime engines, thereby optimizing operational resources and enhancing customer satisfaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional cloud environment resolution techniques are used, then all customer incidents are assigned the same level of priority, but this results in arbitrary selection of resolution order and cost-intensive delays
Solution Approach 1:
The patent changes the parameter of incident priority from a uniform state to a differentiated state based on multiple factors including customer tier, service level agreement, incident severity, and runtime engine criticality. This allows the system to automatically assign different priority levels to different incidents, eliminating arbitrary selection and reducing resolution delays for critical incidents.
Solution Approach 2:
The patent replaces the manual human estimation process for determining resolution order with an automated machine learning system. The trained model automatically classifies incidents into priority categories based on historical data and multiple input features, substituting human judgment with a systematic automated classification mechanism that consistently applies priority rules.
2Extent of automation
If manual estimation and arbitrary selection are used for determining resolution order, then human resources are required for prioritization, but this results in human resource constraints and operational bottlenecks
Solution Approach 1:
The system performs self-service by automatically classifying and prioritizing incidents without requiring human intervention. The machine learning model takes incident data as input and autonomously determines priority levels, enabling the system to manage its own incident resolution queue without human operational overhead.
Solution Approach 2:
The system uses historical incident data and resolution outcomes as feedback to train and improve the machine learning model. This feedback loop allows the system to learn from past performance and continuously refine its prioritization accuracy, improving operational efficiency while maintaining manageable complexity.
3Reliability
If all runtime engines are treated equally regardless of priority, then operational simplicity is maintained, but this results in inability to address critical incidents promptly
Solution Approach 1:
The patent segments the incident population into distinct priority categories based on multiple dimensions including customer tier (enterprise, strategic, standard), service level agreement level, incident severity, and runtime engine criticality. This segmentation allows the system to apply different response strategies to different incident types, ensuring critical incidents receive immediate attention while maintaining a manageable classification structure.
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for automatically classifying cloud infrastructure components for prioritized multi-tenant cloud environment resolution using artificial intelligence techniques are provided herein. An example computer-implemented method includes obtaining historical data pertaining to a multi-tenant cloud environment; training one or more artificial intelligence techniques, using at least a portion of the obtained historical data, for classifying cloud infrastructure components for prioritizing incident-related resolution; classifying one or more cloud infrastructure components, within the multi-tenant cloud environment and associated with one or more server-related issues, into one or more of multiple resolution priority classes; and performing one or more automated actions based at least in part on the classifying of the one or more cloud infrastructure components.