Server Classification Using Machine Learning and Rule-Based Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional server management approaches often lead to improper server decommissioning and over-reclamation, causing outages due to misinterpretation of server activity levels.
Innovation Solution
The use of machine learning techniques for server classification, involving data collection, rule-based analysis, selection of machine learning algorithms, and automated actions based on classification results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional server management approaches are used to determine server activity, then server management simplicity is maintained, but server classification accuracy deteriorates leading to improper decommissioning and over-reclamation
Solution Approach 1:
The system segments the server classification task into multiple components: data collection from various sources, rule-based analysis for initial classification, machine learning algorithms for refined classification, and automated actions. This segmentation allows each component to handle specific aspects of server activity determination, improving overall accuracy while maintaining manageable complexity through modular architecture
Solution Approach 2:
The patent introduces machine learning algorithms as an intermediary between raw server data and final classification decisions. The ML models process and interpret complex server behavior patterns, acting as a mediator that translates raw data into accurate activity classifications, thereby resolving the contradiction between simplicity and precision
2Reliability
If rule-based analysis is used for server classification, then system simplicity is maintained, but classification reliability deteriorates due to misinterpretation of server activity
Solution Approach 1:
The system merges rule-based analysis with machine learning algorithms to create a hybrid classification approach. The rule-based component handles straightforward cases and provides interpretability, while the ML component handles complex patterns and improves reliability. This combination resolves the contradiction by integrating the reliability of ML with the simplicity of rule-based systems
Solution Approach 2:
The patent changes the parameters used for classification from simple threshold-based rules to multi-dimensional features including CPU usage, memory consumption, network activity, and temporal patterns. By transforming the classification parameters into a richer feature space, the system achieves higher reliability while managing complexity through structured parameter selection
3Measurement precision
If machine learning algorithms are applied for server classification, then server activity classification accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by collecting and pre-processing server data from multiple sources before classification. It pre-processes the data to extract meaningful features and prepares it for efficient ML processing. This preliminary action reduces the computational burden during actual classification, thereby improving accuracy while minimizing processing time
Solution Approach 2:
The patent applies partial action by selectively applying machine learning algorithms only to servers that require refined classification, rather than processing all servers uniformly. It uses rule-based analysis for straightforward cases and reserves ML processing for complex or ambiguous cases, thereby achieving high accuracy where needed while reducing overall processing time and resource consumption
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for server classification using machine learning techniques are provided herein. An example computer-implemented method includes obtaining, from at least one data source, data pertaining to server activity attributed to one or more servers; processing at least a portion of the obtained data using one or more rule-based analyses; selecting at least a particular machine learning classification algorithm from a set of multiple machine learning classification algorithms, based at least in part on results from the processing and one or more portions of the obtained data; classifying an activity level of at least a portion of the one or more servers by processing at least a portion of the obtained data using the selected machine learning classification algorithm; and performing at least one automated action based at least in part on results of the classifying.