Neural Network Server Performance Prediction in Edge Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing edge computing systems face inefficiencies in provisioning and monitoring bare metal servers, as they require manual selection of performance data to monitor and often miss analyzing crucial settings and configurations, leading to potential data corruption and performance issues.
Innovation Solution
The implementation of artificial intelligence, specifically a neural network, that analyzes all settings and configuration data of bare metal servers to determine which data to monitor and assess performance criteria, generating a confidence score to predict server performance and notify users of potential issues, while automating the collection and comparison of data against baseline standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If manual selection of performance data to monitor is used in existing edge computing systems, then device complexity is reduced, but measurement precision and reliability of server performance assessment deteriorate
Solution Approach 1:
The system employs trained artificial intelligence that automatically identifies which settings and configuration data to analyze, sets and adjusts weights for the data being analyzed, and determines assessment criteria without requiring manual user selection. The AI autonomously performs the monitoring configuration service, eliminating the need for manual intervention while achieving comprehensive and precise performance assessment.
Solution Approach 2:
The artificial intelligence dynamically changes parameters including the selection of data subsets to monitor, the weights assigned to different settings and configurations, and the criteria for performance assessment. These parameter adjustments are made automatically based on trained models, enabling the system to adapt to different server configurations and performance scenarios without manual reconfiguration.
2Reliability
If comprehensive analysis of all server settings and configurations is performed, then reliability of performance assessment is improved, but loss of time in data processing increases
Solution Approach 1:
The system performs preliminary action by training the artificial intelligence model in advance using historical server data. This pre-trained AI can then quickly assess new server configurations by applying learned patterns and relationships, significantly reducing the time required for performance assessment while maintaining high reliability through the robustness of the training data.
Solution Approach 2:
The artificial intelligence applies partial action by selectively analyzing only the most relevant subsets of server settings and configurations based on weights assigned during training. Rather than uniformly processing all data with equal depth, the AI focuses computational resources on the most predictive features, achieving reliable assessments with reduced processing time.
3Productivity
If automated artificial intelligence analysis is implemented, then productivity of server provisioning is improved, but device complexity increases
Solution Approach 1:
The artificial intelligence system provides multi-functionality by performing multiple tasks: identifying which data to monitor, setting analysis weights, determining performance criteria, detecting drift from baselines, and generating confidence scores. This single universal AI component replaces what would otherwise require multiple separate manual processes, improving productivity while consolidating complexity into a manageable centralized function.
Data Source
AI summary
This disclosure describes systems, methods, and devices related to testing servers provisioned in an edge computing device. An edge computing device may detect that a server has been provisioned to access a public network cloud using backbone routers of the edge computing device; provide a neural network for evaluating a probability that a performance of the server will satisfy performance criteria, the neural network trained based on training data comprising labeled settings data and feature weights; input settings and configurations associated with the provisioning of the server as inputs to the neural network; and generate, using the neural network, based on the inputs and the training data, a confidence score indicative of the probability.