Cloud Hot Spot Detection Using Threshold Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current cloud computing services lack a method to accurately and automatically detect and process local hot spots before they occur, relying on post-solution approaches that are inefficient due to the absence of a clear judgment standard, leading to delayed and manual processing.

Innovation Solution

A method and apparatus that acquire real-time data from node devices and processes within a cloud computing environment, using historical fault data to define and detect local hot spots based on preset processing conditions, allowing for automatic detection and processing, distinguishing between critical and non-critical hot spots.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If simple alarm detection or feedback methods are used for local hot spots, then the system complexity is reduced, but the detection precision and response time deteriorate

Engineering Contradiction:
Improvesystem complexityVSAvoiddetection precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments the local hot spot detection into multiple dimensions: resource usage metrics (CPU, memory, storage, network), time-based analysis (trending data over multiple time points), and threshold-based classification. This segmentation allows precise detection without requiring overly complex unified systems.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary action by pre-defining multiple threshold levels (first threshold, second threshold, third threshold) and pre-configuring processing actions for different hot spot types. This allows the system to automatically respond to detected hot spots without requiring complex real-time decision-making logic.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If no accurate judgment standard is established for local hot spots, then the ease of operation is improved, but the productivity and automated processing capability deteriorate

Engineering Contradiction:
Improveease of operationVSAvoidprocessing efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent transforms the abstract concept of 'local hot spot' into concrete measurable parameters: resource usage thresholds, trending rates, and specific combinations of metrics. By defining clear parameter-based judgment standards (e.g., CPU usage exceeding first threshold combined with memory exceeding second threshold), the system enables automated detection and processing while maintaining operational simplicity.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If manual participation is required for processing local hot spots, then the reliability of processing is improved, but the loss of time and automated processing capability deteriorate

Engineering Contradiction:
Improveprocessing reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements feedback mechanisms where detection results automatically trigger corresponding processing actions based on pre-configured rules. The system monitors resource usage, compares against thresholds, and executes predefined responses (such as alerting, load balancing, or resource allocation adjustments) without requiring manual intervention, thereby reducing processing time while maintaining reliability through systematic feedback loops.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system enables self-service by automatically detecting local hot spots, classifying them according to pre-defined criteria, and executing appropriate processing actions without human intervention. The automated system serves itself by continuously monitoring, detecting, and responding to hot spots using pre-configured rules and thresholds.

Inventive Principle:
Principle #25Self-service

4Device complexity

If post-solution methods are used for local hot spots, then the device complexity is reduced, but the loss of time and business impact deteriorate

Engineering Contradiction:
Improvesystem complexityVSAvoidresponse time
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-configuring multiple threshold levels and corresponding processing actions before hot spots occur. The system continuously monitors resource usage and automatically triggers predefined responses when thresholds are exceeded, enabling proactive rather than reactive handling of local hot spots and reducing business impact.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3832985B1Method and apparatus for processing local hot spot, electronic device and storage medium
Publication Date: 2023.04.05 BEIJING BAIDU NETCOM SCI & TECH CO LTD
  • EP3832985B1 patent drawingFigure 1~2
  • EP3832985B1 patent drawingFigure 3~4
  • EP3832985B1 patent drawingFigure 5~6

AI summary

A method and apparatus for processing a local hot spot, an electronic device and a storage medium. The specific implementation of the method includes: acquiring, at set times, standalone data of each node device and process data of each process within the each node device in a cloud computing environment; detecting, based on the standalone data of each node device and the process data of each process within the each node device, and pre-acquired historical fault data corresponding to each local hot spot, whether each local hot spot satisfies a preset processing condition corresponding to each local hot spot; and processing the local hot spot satisfying the processing condition, in response to detecting that a local hot spot satisfies the corresponding processing condition.