Continuing from the previous article: Software architecture style in the series of knowledge points for software system architect designers (2)
This Eleven is destined to be an Eleven that cannot be relaxed and remains "tight". Since I have signed up for the National Computer Technology and Software Professional and Technical Qualification (Level) Examination, the exam will be held on November 4th, so the 8-day holiday must not be wasted and must be utilized well. Now refine and record each core knowledge point one by one.
Chapter:
Chapter 7. Basic knowledge of system architecture design
Section 3. Software architectural style
2. Segmentation
(3) Data-centric architectural style
Data-centric architectural styles mainly include :
Warehouse architecture style and blackboard architecture style .
- Warehouse architecture style
A warehouse is a central place where data is stored and maintained. In the warehouse style, there are two different components: a central data structure and a set of independent components that operate on the central data .
The central data structure describes the current state of the data; the interaction between the warehouse and individual components can vary greatly in the system.
The connectors of the warehouse architecture style are the interactions between the warehouse and independent components .
- blackboard architecture style
The blackboard architectural style is suitable for solving complex unstructured problems . It can comprehensively use different knowledge sources in the solution process, making it easier to express, organize and solve problems . The blackboard system is a problem solving model and a conceptual framework that organizes reasoning steps , control state data , and domain knowledge for problem solving . It organizes the solution space of the problem into one or more application-related hierarchical structures. Each layer of information in the hierarchical structure is described by a unique vocabulary, which represents a partial decomposition of the problem. Domain- related knowledge is divided into independent knowledge modules , which convert information in a certain level into information at the same level or adjacent levels. Various applications are realized through the combination of different knowledge expression methods, reasoning frameworks and control mechanisms. The biggest factor that affects the design of a blackboard architecture is the nature of the application problem itself, but the blackboard architectures that support applications share many similar features and building blocks.
For specific application problems, a blackboard system can be designed by selecting components from a variety of blackboards (shared data), knowledge sources, and control modules , or by utilizing a pre-customized programming environment for the blackboard architecture.
Traditional applications of blackboard systems are in the field of signal processing, such as speech recognition and pattern recognition . Another application is loosely coupled proxy data sharing access .