Industrial Monitoring System Using Non-Relational Data Warehouse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing industrial monitoring systems face inefficiencies in storing and extracting large amounts of data, particularly for small-to-medium sized enterprises, as they struggle with the time-consuming process of storing and retrieving data from relational databases, which becomes even more challenging with increasing scale and complexity, and the high costs of real-time database systems.
Innovation Solution
An industrial monitoring system architecture comprising a device layer, data sharing layer, data routing layer, data scheduling layer, service layer, and presentation layer, where the data sharing, routing, and scheduling layers are organized using non-relational databases or a mixed relational and non-relational database approach, enabling real-time data processing and efficient data query, with a service layer implemented using domain-driven design and service-oriented architecture, and cloud processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional relational databases are used for data storage and processing, then data can be stored with complex logical relationships, but data storage and extraction become extremely time-consuming when dealing with large amounts of sample data
Solution Approach 1:
The patent introduces a data warehouse that creates a copy of data from the relational database, allowing data mining and analysis operations to be performed on the copied data without affecting the original database. This copying mechanism enables efficient data extraction and processing while preserving the source data integrity.
Solution Approach 2:
The data warehouse acts as an intermediary layer between the relational database and data mining applications. It pre-processes and organizes data in a format optimized for analysis, eliminating the need for time-consuming direct queries on the original database and enabling faster data extraction for reporting and image generation.
2Speed
If real-time database systems are implemented to meet industrial networking and real-time requirements, then real-time data processing capability is improved, but the system cost becomes too expensive for small-to-medium sized enterprises
Solution Approach 1:
The patent employs open-source software components and standard hardware configurations to build the data warehouse and processing system, replacing expensive proprietary real-time database systems. This approach achieves real-time processing capabilities through cost-effective technologies, making the system accessible to small-to-medium enterprises.
Solution Approach 2:
The data warehouse is designed to serve multiple functions including data storage, preprocessing, mining, reporting, and image generation within a single integrated platform. This multi-functionality eliminates the need for separate expensive specialized systems while maintaining real-time processing capabilities across various industrial applications.
3Quantity of substance
If large amounts of sample data are added to databases for data mining and scientific calculations, then data analysis capability is improved, but the time required to store and extract this colossal data increases significantly
Solution Approach 1:
The system performs preliminary data processing and organization by pre-loading sample data into the data warehouse in optimized formats. This preliminary action includes data cleaning, transformation, and structuring operations that are completed in advance, enabling rapid extraction and analysis when data mining or scientific calculations are initiated without time-consuming processing delays.
Data Source
AI summary
An industrial monitoring system is disclosed. In an embodiment it includes a device layer, a device driving layer, a data sharing layer, a data routing layer, a data scheduling layer, a service layer, and a presentation layer. The data sharing layer, the data routing layer, and the data scheduling layer are organized by using a non-relational database, or organized in a hybrid way by using the non-relational database and a relational database. The industrial monitoring system disclosed can implement processing such as real-time storage and sharing of a large amount of data, and data mining and retrieval.