ML-Based Impact Hierarchy for Real-Time Data Collection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data collection methods in data center environments are delayed, leading to inaccuracies in determining the root causes of component failures due to changes in component states before data is collected, resulting in discrepancies.

Innovation Solution

The implementation of machine learning algorithms to quickly identify impacted components and collect operational data in real-time upon a failure event, using a component management platform that determines the impact hierarchy and collects data from both the affected and impacting components.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If operational data is collected after a failure event using current approaches, then data collection can be performed, but there is a lapse of time leading to changes in component states and inaccuracies in determining root causes

Engineering Contradiction:

Improveaccuracy of operational dataVSAvoidtime delay in data collection

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system pre-identifies components that may be impacted by a failure event using machine learning algorithms before data collection occurs. By determining the impact hierarchy in advance and collecting data from all potentially impacted components simultaneously upon event detection, the system eliminates time delays and ensures data accuracy reflects the actual state at the time of failure.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If operational data is collected from multiple components simultaneously, then accuracy is improved, but the complexity of identifying which components are impacted increases

Engineering Contradiction:

Improveaccuracy of root cause determinationVSAvoidcomplexity of component impact identification

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system introduces a machine learning-based impact hierarchy model as an intermediary between failure detection and data collection. This model automatically determines which components are impacted by a failure event and establishes their hierarchical relationships, enabling targeted data collection from the correct components without manual analysis or excessive complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system replaces manual or rule-based component impact identification with machine learning algorithms. The ML models analyze component relationships, failure patterns, and system topology to automatically determine the impact hierarchy, substituting complex mechanical or procedural analysis with intelligent computational methods that scale better and provide more accurate results.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11663102B2Event-based operational data collection for impacted components

Publication Date: 2023.05.30 DELL PROD LP

AI summary

A method comprises receiving a notification of an issue with at least one component of a plurality of components in a computing environment. One or more machine learning algorithms are used to determine one or more components of the plurality of components impacted by the issue with the at least one component. The method further comprises collecting operational data for the at least one component and the one or more impacted components.