Classifier for System Management Policy Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional system management tools for cloud infrastructure lack a broad view of situations impacting operating performance, relying on local knowledge and requiring significant human intervention for policy decisions.
Innovation Solution
A system management tool that receives situational classification results from a classifier, allowing it to take automatic management actions based on a broader understanding of the system's operating conditions, including external factors, thereby enhancing decision-making and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If system management tools base policy decisions solely on local knowledge, then device complexity is reduced, but the ability to respond to high-level situations impacting operating performance deteriorates
Solution Approach 1:
The patent introduces an intermediary component (the classifier) that sits between the managed system and the system management tool. This classifier receives situational data, processes it through machine learning models, and outputs simplified situation classifications that the management tool can understand and act upon. This intermediary resolves the contradiction by enabling the management tool to access high-level situational awareness without increasing its own internal complexity.
Solution Approach 2:
The patent segments the situational analysis function into a separate, dedicated component (the classifier) rather than embedding it within the system management tool. This segmentation allows the management tool to maintain simplicity while still benefiting from comprehensive situational analysis performed by the specialized classifier component.
2Ease of operation
If conventional system management tools require significant human intervention for policy decisions, then ease of operation is improved, but productivity deteriorates
Solution Approach 1:
The patent implements self-service automation where the system management tool automatically executes policy decisions based on situation classifications without requiring human intervention. The machine learning classifier continuously monitors system states and triggers appropriate management actions autonomously, allowing the system to manage itself and eliminating the need for manual policy implementation while maintaining operational simplicity.
Solution Approach 2:
The patent establishes a closed-loop feedback system where the classifier continuously monitors system operating conditions, classifies situations, and feeds this information back to the system management tool, which then automatically adjusts policies. This continuous feedback loop enables automated, real-time decision-making that improves productivity while keeping the system easy to operate through centralized automated control.
Data Source
AI summary
An information processing system comprises a system management tool, a managed system and a classifier. The system management tool receives a result of a situation classification operation from the classifier, and automatically takes a management action relating to the managed system based at least in part on the result of the situation classification operation. The situation classification result may be determined by monitoring operation of the managed system in a situational state space having dimensions that characterize operating conditions of the system, and identifying a current state of the system in the situational state space. The management action automatically taken by the system management tool may comprise, for example, setting at least one policy for the managed system by selecting and deploying a particular one of a plurality of policy sets responsive to the result of the situation classification operation.