Differences between real-time/historical databases and relational databases

Differences between real-time historical databases and relational databases 1. Introduction to real-time databases and relational databases (1) real-time database is a branch of the development of database systems, it is suitable for processing constantly updated and rapidly changing data and transactions with time constraints. Real-time database technology is a combination of real-time systems and database technology. Real-time

Differences between real-time/historical databases and relational databases 1. Introduction to real-time databases and relational databases (1) real-time database is a branch of database system development, it is suitable for processing constantly updated and rapidly changing data and transactions with time constraints. Real-time database technology is a combination of real-time systems and database technology. Real-time

Differences between real-time/historical databases and relational databases

1. Introduction to real-time databases and relational databases

(1) Real-Time Database

Real-time database is a branch of the development of the database system. It is suitable for processing constantly updated and rapidly changing data and transaction processing with time constraints. Real-time database technology is a combination of real-time systems and database technology. Real-time databases first emerged based on advanced control and optimization control, and have high real-time data requirements. Therefore, real-time, efficient, and stable are the most critical indicators of real-time databases.

The concept of early real-time databases is what we call memory databases. They act as the data hub and effectively integrate factory-level isolated DCS data, in plant-level applications, the data of a DCS can provide data support for other DCS process algorithms, which effectively solves the problem of data islands and expands the functions of DCS, real-time Databases play an important role in advanced control and optimization control.

However, early memory databases cannot effectively solve key issues such as fine-time compression and storage of real-time data, highly classified and abstract Numerical Attributes of industrial model objects, and efficient real-time retrieval and processing of large-capacity data. Real-time databases are specialized and productized in data communication, data organization, data storage, data retrieval, data access, data processing, and data presentation, it provides convenient and stable data support for building analysis applications based on large-capacity real-time historical data, so that the application system can make full use of valuable real-time historical data from a higher and deeper level.

Currently, real-time databases are widely used in electric power, petroleum and petrochemical, transportation, metallurgy, military, environmental protection, and other industries. They are used to build industrial production scheduling and monitoring systems and command systems, an indispensable basic software for producing real-time historical data centers.

(2) relational databases

A relational database is a database that uses a relational model to organize data. Relational Model was first proposed by IBM in 1970. In the next few decades, the concept of relational model has been fully developed and gradually become the mainstream model of database architecture. In short, a relational model refers to a two-dimensional table model, and a relational database is a data organization composed of two-dimensional tables and their relationships. Relational databases have the following features:

Easy to understand: the structure of a two-dimensional table is very close to the logic world. The relational model is easier to understand than other models such as mesh and hierarchy.

Easy to use: the general SQL language makes it very convenient to operate relational databases. programmers and even data administrators can operate databases logically without having to understand the underlying implementation.

Easy to maintain: Rich integrity (entity integrity, reference integrity, and user-defined integrity) greatly reduces the probability of data redundancy and data inconsistency.

Currently, relational databases are widely used in various industries to build management information systems, such as ERP, MIS, and EAM. They are used to build management information systems, basic software is indispensable for storing and processing relational data.

Ii. Differences between the two databases

(1) Data Organization

Real-time databases can be easily understood as such databases: they are composed of three databases: The measurement point information database, the real-time database, and the historical database.

A real-time database is a memory snapshot database that reflects the changes in the time stamp, value, and quality of real-time production data in seconds. You can query the real-time data values (values, timestamps, and quality) of real-time production data from this database ).

The historical database contains a table that uses the measurement point Name field and the time field as the keyword. Another important field in this table is the numerical field, which is used to store the collection value of the measurement point, in addition to these fields, they can also contain data status and data quality fields. As time changes, the real-time data in the real-time database is constantly compressed and filtered, and the data in the table in the disk's historical data file is updated. You can query historical sample values or historical interpolation data of real-time production data from this database.

Relational databases design data tables based on the relationship between entities.

(2)System Stability

Because real-time Databases record production-related data and time-related data, they must be able to run stably for a long time. Otherwise, data loss may occur. Currently, some real-time databases already have the data caching function. When the communication between the data collection machine and the real-time database server fails, the collected data can be cached locally, when the communication returns to normal, the cached data is written to the real-time database server, which greatly guarantees data integrity.

For relational databases, if they are not applied to key businesses, such as financial securities, the stability requirements are generally not very high.

(3)Data Source

For different types of enterprises, the data sources of real-time databases are also different. The main sources include DCS, SCADA, manual input, and relational databases. The main feature of the data is that it is directly related to production, and most of the data is numerical data, such as the pressure, temperature, flow, location, voltage, current, power of the device or medium.

The source of relational databases is more diverse. In addition to numerical data, descriptive data, such as the name, home address, and other information are also recorded.

Generally, the data source of a real-time database is a device. Relational databases generally come from people.

(4)Data Compression

Real-time Databases collect a large amount of data, such as 10000 points of data every five seconds. Assume that all data is collected with 32-bit floating point numbers, the data size of a day (only the value attribute does not include the time attribute) is 10000 * (60/5) * 60*24*4/1024 = 675000 K, which is about MB. It can be seen that the data volume is huge and the disk space is large, and the access speed to the data is also reduced. Therefore, most database manufacturers have developed their own data compression algorithms to compress data. Common compression algorithms include lossless compression, lossy compression, and secondary compression. Among them, lossless compression is generally based on general compression theory and adopts typical compression algorithms such as huffman, while lossy compression takes into account the characteristics of industrial real-time data, the second-level compression technology uses the two data compression technologies at the same time. Lossless compression of real-time databases is based on general compression theory. Currently, the famous lossy compression algorithms include common revolving door compression algorithms and some flexible compression algorithms (for example, based on the revolving door algorithm, the quadratic mean variance is used for deviation comparison to improve data restoration accuracy ), the principles of these algorithms are relatively simple.

Relational databases do not compress data.

(5)Data Access Method

Real-time databases generally access data in the following three ways:

1) Use your own API. This method is the most efficient and simplest.

2) ODBC or OLEDB is not commonly used, mainly for users who are new to real-time databases or who have previously known about relational databases.

3) use Opc to access data. Opc is a widely used industrial standard. Although its efficiency is not high, many manufacturers currently support it.

Relational databases access data through structured query languages (SQL.

(6)Application fields

Process Control System is one of the most important application fields of real-time database systems. during the operation of the production device, the real-time database records the operation data of the Collection Device in real time, keeps abreast of the running status of the device, and monitors and analyzes key data in the production process, timely handling of problems, ensuring the safety and stability of production operations, and timely reflecting changes in production conditions; through monitoring and analyzing the process that affects the amount of raw materials and the amount of water, electricity, and steam, problems can be detected in a timely manner. Especially for production scheduling personnel, the material supply can be balanced in a timely manner to reduce unit consumption, improve economic efficiency.

Relational databases are widely used and can be seen in all walks of life. Most applications are used in management, such as management information system (MIS) and Customer Relationship Management (CRM.

(7)Client tools

Since there are no uniform standards for real-time databases, all real-time database client tools are provided by the vendor, mainly data presentation tools, configuration tools, and management tools. Because it is mainly for the process industry, real-time databases are equipped with configuration and release tools.

Relational databases are widely used and provide a wide range of customer tools. In addition to the tools provided by the vendor, third-party software vendors can also develop client tools for relational database vendors.

Iii. integration trend between two databases

Although there are many differences between real-time databases and relational databases, the current trend of real-time database and relational database integration is becoming more and more obvious. Integrating the relational databases and real-time Databases Used in the production management information system can meet the control and management requirements at the same time and truly become an integrated platform for management and control. For example, most industries currently use real-time databases and relational databases as basic database software to build basic databases for monitoring and data centers. Real-time Databases process and store real-time production data, and relational databases process and store business relationship data.

Http://jianwei.cto.csdn.net/Article.aspx? Name = jianwei & pointtid = 4611