Cloud Service Interdependency Detection via Performance Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional topology discovery tools for cloud computing systems are intrusive, resource-intensive, and ineffective in detecting performance interdependencies between components, especially in dynamic and scalable environments where network components can create opaque connections, leading to inaccurate system management and potential failures during reconfiguration or maintenance.
Innovation Solution
A computer system that generates topological maps of networked computing resources by analyzing performance data using principal component analysis (PCA) and correlation analysis to identify performance interdependencies, creating an undirected graph that visually represents these relationships, allowing for indirect detection of connections and continuous updating of system topology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional topology discovery tools are used to detect network connections, then system topology can be mapped, but the tools are intrusive, resource-intensive, and ineffective in detecting performance interdependencies
Solution Approach 1:
The patent replaces traditional mechanical/intrusive topology discovery tools with a data-driven approach using machine learning algorithms. Instead of actively probing and mapping network connections through intrusive tools, the system analyzes existing performance data to infer topological relationships and performance interdependencies, thereby eliminating the need for resource-intensive discovery mechanisms while improving detection accuracy
Solution Approach 2:
The patent introduces performance data as an intermediary medium to detect topological relationships. Rather than directly probing network connections, the system uses performance metrics (CPU utilization, memory usage, network bandwidth, etc.) as intermediaries to indirectly infer the topology and performance interdependencies between cloud computing resources, achieving non-intrusive detection
2Adaptability or versatility
If cloud computing resources are dynamically reconfigured to meet fluctuating demand, then service scalability is improved, but accurate system management becomes difficult and potential failures occur during reconfiguration
Solution Approach 1:
The patent implements continuous feedback mechanisms by constantly analyzing performance data from cloud computing resources and updating the topological map in real-time. This feedback loop enables the system to adapt to dynamic reconfigurations, maintain accurate awareness of performance interdependencies, and prevent failures by identifying critical dependencies before reconfiguration operations are executed
Solution Approach 2:
The patent performs preliminary analysis of performance interdependencies and topological relationships before reconfiguration operations. By pre-identifying critical performance dependencies and potential failure points through continuous monitoring and machine learning analysis, the system can plan reconfiguration operations to maintain reliability while achieving scalability
3Measurement precision
If performance data from multiple categories is collected for analysis, then detection accuracy of interdependencies is improved, but data processing complexity increases
Solution Approach 1:
The patent segments the complex task of detecting performance interdependencies into distinct processing stages: data collection from multiple performance categories, data preprocessing and normalization, feature extraction, machine learning model analysis, and result interpretation. This segmentation allows each stage to handle specific aspects of the data, reducing overall processing complexity while maintaining high detection accuracy through specialized processing at each stage
Solution Approach 2:
The patent transforms raw performance data from multiple categories into standardized parameters and features that are suitable for machine learning analysis. By changing the parameters through normalization, aggregation, and feature engineering, the system reduces data complexity while preserving the essential information needed for accurate detection of performance interdependencies
Data Source
AI summary
A computer system includes a processor, a memory, a data collector, a relationships analyzer, and a topological map generator. The data collector retrieves performance data in a specific set of performance categories for computing resources in a computing system for a time interval. The relationships analyzer, for each computing resource-to-computing resource pair in the computing system, performs a correlation analysis of the respective behavior values of the computing resources in the pair, and identifies the computing resource-to-computing resource pairs that have correlation values exceeding a pre-determined threshold level as having performance interdependencies. The topological map generator prepares an undirected graph of the computing resources that have performance interdependencies, and displays the undirected graph as a topographic map of the computing resources in the computing system.