Impact Factor: isra (India) = 971 isi (Dubai, uae) = 829 gif

4. DATA DESIGN 
When designing data-driven systems, the inputs 
and outputs are the most important, not processing and 
computation. Therefore, the order of analysis and 
design is somewhat different from the traditional order 
when developing, say, real-time systems. Namely: 
data structures are defined first, procedural 
components are built as derived from data structures. 
In fact, the design process consists in defining data 
structures, merging them into the prototype of the 
program structure and filling this structure with 
detailed data processing logic. 
Jackson's structural design is a classic example 
of this approach. Its basic design procedure is intended 
for "simple" programs (a "complex" program is 
broken down into simple ones using traditional 
methods) includes the following 4 stages: 
- Data design phase. 
- Program design stage. 
- Operations design phase. 
- The stage of designing texts. 
At the design stage of data, it is necessary to 
define the structures of input and output data and, on 
their basis, build some model of the internal data 
representation. This requires: 
Build a model of the functional requirements of 
the system at the level of control processes (analyzing 
the subject area). 
Then build a data model at the entity-relationship 
level (by analyzing processes). 
DATA FLOW DIAGRAMS 
Data flow diagrams (DFDs) are the most popular 
tool for building a system model at the level of control 
processes. With their help, these requirements are 
broken down into functional components (processes) 
and presented as a network connected by data streams. 
The main goal is to demonstrate how each process 
transforms its inputs into outputs, as well as to reveal 
the relationships between the processes. DFDs are 
represented using Jordan notation. 
The following concepts are basic: 
Data stream. A mechanism for transferring 
information from one part of the system to another. 
They are depicted by named arrows, the orientation of 
which indicates the direction of movement of 
information. 
Process. Produces output streams from input 
streams according to some rule. It is depicted as a 
circle, inside which the name of the process and its 
number are placed (for links to it inside the diagram). 
Data storage (storage). Allows you to save data 
between processes. The information it contains can be 
used at any time after it has been defined, and the data 
can be selected in any order. It is depicted by two 
parallel horizontal lines, between which the name of 
the repository is written. 
External entity. An entity outside the context of 
the system, which is a source or receiver of data (does 
not participate in data processing). It is represented by 
a rectangle with a name inside. 
When constructing diagrams, it is advisable to 
observe the following sequence: 
Dividing many requirements into groups. 
Identification of external and internal system 
objects. Extract information about objects from the 
requirements for primary (input) and secondary 
(output) documents. 
Identification of the main types of information 
circulating between the system and external objects. 
Development of a common DFD. At the same 
time, proceed from the recommendation: place from 3 
to 6-7 processes on each diagram (the limit of human 
capabilities of simultaneous perception). Combine 
more complex processes into one process, and group 
threads. 
Formation of DFD of lower levels by 
decomposing complex processes into parts (in parallel 

---