DATA FLOW DIAGRAMS (DFD)
The DFD is one of the tools of modern structured analysis, most important for the analysis of graphic models, which allows visualizing a system as a network of functional processes connected to each other by channels (data flow) and data storage repositories.
These diagrams allow us to see how data flows through the organization, the processes and transformations that such data undergo, and the different types of outputs.
COMPONENTS AND SYMBOLS USED
- Data Flow Process Warehouse Terminator
ELEMENTS OF THE DATA FLOW DIAGRAM
PROCESSES
Generally, the process (function or transformation) is represented by a circle or by a bubble, and they are actions taken on the data, such as, Calculate, Compare, Print, point, mark, authorize, store, validate, inform, produce, others. The processes show a part of the system that transforms inputs into outputs, that is, it shows how one or more inputs is transformed into one or more outputs.
NAME OF PROCESS
The name of a process consists of a VERB-OBJECT phrase, and describes what it does; for example:
CALCULATE-TAX
AUTHORIZE- SIGNATURE
AUTHORIZE-INVOICE
AUTHORIZE-PURCHASE-ORDER
VALIDATE- SUPPLIER
GENERATE REPORTS
Also, the processes can be described (although it is not recommended) with the name of a person or a group of people, a computer or a mechanical device, in any case the key word is "Who" or "What" is performing it.
CONSIDERATIONS ON DATA FLOW.
DATA FLOW.
They are labeled vectors or arrows, or simply lines with directional notation, that show the content of what enters or leaves a process. In addition, they show the movement of blocks or packets of information from one place in the system to another. The tip of the arrow indicates the destination or origin of the data.
- They must be labeled or named with the data they carry, except when they leave or enter a warehouse, since they describe what they contain. However, if only one instance is extracted it must be tagged.
SURVEY PURIFICATION SYSTEM
- The data that moves along the flow can travel from one process to another (as input), to a warehouse or to a Terminator (source or destination of the data). The flow carries only one type of data packet as its name indicates, but there are exceptions, grouping elemental flows into one. Example: File = CLIENTS
CUSTOMER NAME; ADDRESS-CLIENT; BALANCE-CLIENT; MAXIMUM-CREDIT
The other type of warehouse is the one that is established for convenience; for example, the Orders warehouse described below:
(See PDF)
1. Both processes are expected to run on the same computer, but there is not enough memory to cover both at the same time. Thus, the ORDERS warehouse is created as an intermediate file, since the available implementation technology has forced the processes to be executed at different times.
2. Either or both processes are expected to run in a hardware configuration that is unreliable. Thus, the ORDERS warehouse is created as a backup in case any of the processes is aborted.
3. Different programmers are expected to implement the two processes. Thus, the ORDERS warehouse is created to test and correct, so that if the complete system does not work, both groups can see the contents of the warehouse and detect the problem.
4.The analyst or designer thought that the user could someday make access to the ORDES warehouse for some other reason, even when he has not expressed such interest. In this case, the warehouse is created anticipating future needs of the user.
WAREHOUSE CHARACTERISTICS
- The name used is the plural of the name used for data packets entering and leaving the warehouse via streams.
BOOKS
- A warehouse should not be referred to as a physical storage device (file or database; for example, a tape file or an organized file with IMS, DB2, ADABAS, IDMS, or some other database management system)), something commonly practiced by experienced analysts. In most cases in a warehouse, flows into or out of it are not tagged unless a portion of it is removed. A warehouse is passive and the data is not. will travel along the flow. A flow to a warehouse can be described as a write, update, or delete:
- One or more new packages are being stored One or more packages are being modified or changed One or more packages are being removed from the warehouse
TERMINATORS
External and internal terminators or agents, as they are also known, are the sources or destinations of the data. Normally, an agent is considered external when it is clearly external to the company, some examples of these are: Clients, Suppliers and Government Agencies. The agents are internal, when they refer to tasks carried out within the company but that are not part of the system; however, they supply inputs to or receive outputs from it. Other departments, employees, or information systems may be cited as internal agents. Internal agents can also encompass the end users of a system, who are often sources of inputs (data) and destinations of outputs (information).
TERMINATOR CHARACTERISTICS
- Their name must come in capital and singular. They are external to the system, the flows that connect them to a process or to a warehouse represent the interface between the terminator and the rest of the world. Those responsible for analysis or design cannot change their content or the way they work. Therefore, the model being developed must be flexible enough to allow the designer to choose the best implantation. In this sense, the analyst cannot modify the content, the organization, or the internal procedures of the terminators. The relationships that exist between the terminators are not shown in the DFD graph, since by definition they are external to the organization. If it were the case that the relationship existed, and is of interest to the analyst,then terminators would be part of the system and should be modeled as processes.
DATA FLOW DIAGRAM
STEP-BY-STEP PROCESS MODELING
STEPS FOR THE CONSTRUCTION OF DATA FLOW DIAGRAMS STEP 1
Develop a CONTEXT data flow diagram - it places the system within an environment context; that is, how the system interacts with other systems and with the company considered as a whole. Defines the scope and limits of the system and the project.
When drawing a context diagram:
- Use a single command prompt. Label the command prompt so that it represents the entire system. A verb plus an object can be used. Do not number the process symbol. Include all terminators in the system. Show all data flows between terminators. Strategies for determining them: Ask your end users what events or transactions the system should respond to. For each event, ask your end users what responses the system should produce. Ask what the fixed-format reports are. to be produced by the system. Identify the net data sources for each event. Identify the net containers for each response or output that the system should generate. Identify all possible external data stores.Draw a context diagram for all the above information.
DATA FLOW DIAGRAM FIGURE N
- STEP 2: Prepare a General Data Flow Diagram (Figure 0)
If a system needs to be documented in greater detail than the Figure 0 diagram, one or more Figure n diagrams can be used. A diagram in Figure n documents a single DFD process in greater detail. The n represents the number of the next highest level process being documented
- Guide for its construction.
- Its construction begins once its components are known, which must be identified jointly with the users. Choose meaningful lasting names for the components. Number the processes so that they serve as a reference for the analyst for their subsequent explosion. Avoid excessively complex DFDs.Maintain consistency between processes and other models.
¨ Avoid infinite sinks - bubbles that only have inputs but no outputs.
¨ Avoid spontaneously generated bubbles - they are bubbles with only exits.
¨ Be careful with unlabeled flows and processes.
¨ Be careful with write-only or read-only stores - all stores must have both inputs and outputs, except the external store that serves as an interface between the system and some external terminator. 6. Restrict a single DFD to no more than six or eight processes
- A page must be used for a particular DFD.
DATA FLOW DIAGRAM
DATA FLOW
- They represent entering data into a process or obtaining data from a process. They represent updating data in a file, database, or other means of data storage. It is a pathway through which composition data packets transit. data can travel by any given route.
TERMINATORS (SOURCES OR DESTINATIONS OF DATA)
- Internal agents (inputs to the system or sources), refer to tasks carried out within the company, but which are not part of the scope of the system and also provide inputs or receive outputs from it. Examples: other departments, employees or information systems. External Agents (exits from a system or destinations) are those that are clearly outside the company. Examples: customers, suppliers and government agencies.
DATA WAREHOUSES
A warehouse is a data inventory and they describe things about which the company wants to store data. It is the most common point of union between data models and process models.
CONTEXT DATA FLOW DIAGRAM FOR A DISTRIBUTION SYSTEM
DATA FLOW DIAGRAM OF FIGURE 0- DISTRIBUTION SYSTEM
Explosion of the Data Flow Diagram.
So far we have seen representations of the DFD at a very high or managerial level of the system, but the user might ask, for certain details that are not shown at that level.
The idea of the DFD explosion; it is to organize the global DFD into a series of levels, in such a way that each one successively provides more details about a portion of the previous level. As we mentioned, the first diagram is called the Context Diagram, which represents the complete system. Data streams show the interfaces of the system with its environment (terminators). These terminators, in turn, can be connected to external warehouses that may exist.
The DFD, which follows the Context Diagram, is known as the Figure 0 Diagram, and represents the main functions of the system as well as its main interfaces.
Thus, successively we are exploiting until representing the level of detail required by the User.
RULES FOR EXPLOSION OF DATA FLOW DIAGRAMS
- THE FIRST LEVEL OF THE DFD ONLY PRESENTS A GENERAL SCHEME OF THE SYSTEM. THE COMPLEX OF THE LOWEST-LEVEL DFD SHOWS A DETAILED VIEW OF THE SYSTEM. THE ONLY DIFFERENCE BETWEEN THE DFD OF EACH LEVEL IS VERY CONVENIENT IN THEM IDENTIFY EACH OF THE DIAGRAMS THAT EXPLOIT
A PROCESS WITH THE BUBBLE NUMBER THAT CORRESPONDS TO IT IN THE FATHER DIAGRAM. INSIDE AN EXPLOSION DIAGRAM THE BUBBLES IT CONTAINS SHOULD ALSO BE NUMBERED, IN SUCH A SIMPLE WAY TO IDENTIFY THE DFD PARENTS AND ALL THEIR CORRESPONDENTS DEF DERIVED.
RULES FOR EXPLOSION OF DFD (Continued)
DFD EXPLOSION (Continued)
EXPLOSION OF DFD
FIRST EXPLOSION
FIRST EXPLOSION (FIGURE 1) OF THE MERCHANDISE DISTRIBUTION SYSTEM
DATA FLOW DIAGRAM FIGURE 2 - ORDER STOCKS OF
INVENTORY SYSTEM FIGURE 1.2 DIAGRAM
CONTEXT DIAGRAM OF THE EMISSIVE TOURISM SYSTEM
FIRST LEVEL DATA FLOW DIAGRAM
STATE ADDRESSES
EMISSIVE TOURISM SURVEY
DATA FLOW DIAGRAM - EMISSIVE TOURISM HEADQUARTERS PROCESSES (DEE)
Download the original file
