Edge Server Analytics Engine Cloud Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current IoT systems face challenges in achieving accurate and real-time analytics due to latency issues caused by data transfer between IoT devices and cloud servers, and the limitations of edge devices in processing power and access to new training data, which hinder the adaptability and responsiveness of IoT networks.
Innovation Solution
A method is introduced where an analytics engine hosted by a cloud server assists in training an edge server's analytics engine, allowing for improved prediction accuracy by leveraging cloud resources while reducing latency and enhancing edge device capabilities through automated training and data management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is transferred between IoT devices and cloud servers for analytics, then comprehensive data processing is achieved, but latency increases and real-time responsiveness deteriorates
Solution Approach 1:
The system segments analytics processing between cloud servers and edge servers. The cloud server handles comprehensive data processing for high accuracy analytics, while the edge server handles time-critical processing for real-time responsiveness. This segmentation allows each component to optimize for its specific function, resolving the contradiction between accuracy and latency.
Solution Approach 2:
The edge server acts as an intermediary between IoT devices and the cloud server. It receives data from IoT devices, performs preliminary analytics processing locally to maintain real-time responsiveness, and selectively transfers only necessary data to the cloud server for comprehensive analysis. This intermediary role reduces latency while maintaining analytics accuracy.
2Speed
If edge devices are given more processing power to reduce latency, then real-time responsiveness improves, but device complexity and resource requirements increase
Solution Approach 1:
The edge server performs partial analytics processing locally rather than requiring full processing capability. It handles time-critical analytics tasks that require low latency, while less time-sensitive tasks are deferred to the cloud server. This partial action approach improves real-time responsiveness without requiring excessive processing power at the edge.
3Measurement precision
If edge devices have access to more training data to improve model accuracy, then prediction precision improves, but access to new training data becomes more difficult due to isolation
Solution Approach 1:
The system implements a feedback mechanism where the edge server sends its predictions and local training data to the cloud server. The cloud server uses this feedback to generate updated training data and refined analytics models, which are then transferred back to the edge server. This feedback loop enables continuous improvement of prediction accuracy while maintaining the edge server's relative isolation.
Data Source
AI summary
A method for training an analytics engine hosted by an edge server device is provided. The method includes determining a classification for data in an analytics engine hosted by an edge server and computing a confidence level for the classification. The confidence level is compared to a threshold. The data is sent to a cloud server if the confidence level is less than the threshold. A reclassification is received from the cloud server and the analytics engine is trained based, at least in part, on the data and the reclassification.