DFD diagram of software requirements

DFD graph is a ing method based on the closeness of data and data. It is a structured analysis method from the problematic space to a certain representation. DFD diagram plays an irreplaceable role in software requirement analysis. DFD diagram is mandatory in the soft examination, and is also a very important diagram in the software engineering. The following is a summary of my online materials and my knowledge.

Category

A Data Flow Process is represented by a box, line, and deformation symbol.

One is to represent the data flow process with circles and connection arcs as their basic symbols.

 

Painting Method

From top to bottom, expand layer by layer; balance input and output.

We can see that the data flow chart is hierarchical, and the method of top-down layer-by-layer decomposition is used for drawing.

First, draw a top-layer (0-layer) data flow diagram. There is only one top-layer data flow chart, which describes the system's overall processing functions, input and output.

The next step is to break down "processing" in the top-layer data flow chart, that is, to break down "accounting processing" into more "processing ". The 2nd images are one of the data flowcharts at the first layer after the processing is decomposed in The 0th layer.

The number of layers of a Data Flow Chart depends on the actual situation. For a complex large system, it can be divided into seven or eight layers. To improve normalization, it is necessary to number each element in the graph. Generally, a letter is given at the beginning of the number to indicate different elements. P can be used for processing, D is used for data flow, F is used for data storage, and S is used for external entities.

 

Features

Abstract. In the data flow diagram, the specific organization, workplace, personnel, and material flows have been removed, leaving only the data storage, flow, processing, and usage. This abstraction can summarize the internal regularity of information processing.

General. It connects the system's processing processes for various businesses to form an overall structure.

 

Composition of a data flow chart

 Data Stream:Arrow. Is composed of a set of fixed data, indicating the data flow. It is worth noting that the data flow diagram describes the data flow rather than the control flow. In addition to data flowing to or from data storage, you do not need to name the data flowing out. Each Data Stream must have a proper name to reflect the meaning of the data stream.

Processing:Elliptical representation. Processing describes the transformation between the input data stream and the output data, that is, after which the input data stream is converted into the output data. Each processing task has a name and a number. The number indicates the layer and graph of the data flow graph in which the processing is located, and shows which processing is used to separate the data flow graph.

Data storage:Double underline. Data storage indicates the data temporarily stored. Each data storage has a name. Generally, it is a table in a database file.

Data Source point endpoint (external entity ):Rectangle. An external entity is a person or organization that exists outside the software system. It points out the source of the data required or the location of the data produced by the system. Generally, it only appears on the top layer of the data flow diagram.

 

Data Flow Diagram Design Method

1. input and output of the painting subsystem: regards the entire system as a big processing, where the data comes from and where it goes

2. Inside the painting subsystem: Process top-level Graphs into several processes and connect them with data streams.

3. Inside the image processing system: each processing system is regarded as a small system, and the input and output data streams are processed as input and output streams of a small system.

4. Draw a sub-processing breakdown chart: Repeat the decomposition process of step 3 for each processing in the DFD diagram decomposed in step 3 until it cannot be further decomposed.

5. Correct correspondence between data flow charts and processing numbers:

 

Notes
1. Naming: No matter data streams, data storage, or processing, Proper naming makes it easy for people to understand its meaning.

2. Draw data streams rather than control flows: Data Streams reflect what the system "does" and do not reflect "How to Do". Therefore, the data stream names on the arrows can only be nouns or noun phrases, the entire figure does not reflect the processing execution sequence.

3. Generally, do not draw material streams: Data Streams reflect data that can be processed by computers, rather than physical objects. Therefore, do not draw material streams for data streams of the target system.

4. Each processing has at least one input data stream and one output data stream, reflecting the source and processing result of the processed data.

5. Number: If a data flow graph is processed into another data flow graph, the upper layer graph is the parent graph and the lower layer graph is the subgraph. The subgraph and all the processing operations should be numbered.

6. Balance between the parent and subgraphs: the input and output data streams of the subgraphs must be consistent with the input and output data streams processed by the parent graph. This is the balance between the parent and subgraphs.

7. Local Data Storage: When data storage in a data flow graph at a certain layer is not an external interface processed in the parent graph, it is only a data interface between some processes in this graph, this is called local data storage.

8. Improve data flow chart Readability: note that a data flow chart should be decomposed into several sub-processes with relatively independent functions to reduce the number of input and output data streams between processes, increase the comprehensibility of data flow graphs.

 

Example

DFD diagram of a hospital ward Monitoring System