Systems Analysis and Design

Data Storage Design

Prepared by:
Dr. Muhammad Iqbal Hossain
Assistant Professor, Dept. of CSE
Brac University

11-1
 Chapter Outline
• Data storage formats.
- Files.
- Databases.
• Moving from logical to physical data models.
• Optimizing data storage.

11-2
 INTRODUCTION
• The data storage function is concerned with how
data is stored and handled by programs that run the
system.
• Data storage design is to
- select the data storage format;
- convert the logical data model created during
analysis into a physical data model to reflect the
implementation decision;
- ensure that DFDs and ERDs balance; and
- design the selected data storage format to optimize
its processing efficiency.
11-3
 Files
• A data file contains an electronic list of
information that is formatted for a
particular transaction.
• Typically, files are organized sequentially.
• Records can be associated with other
records by pointers.
• Sometimes files are called linked Lists
because of the way the records are linked
together using pointers.
11-4
File based DB: appointment
record

11-5
Example of
database:

11-6
 Databases
• There are many types of databases:
- Relational database
- Object database
- Multidimensional database

11-7
 Relational Databases
• The relational database is the most popular kind of
database for application development today.
• A relational database is based on collections of tables,
each of which has a primary key.
• The tables are related to each other by the placement of
the primary key from one table into the related table as a
foreign key.
• Most relational database management systems (RDBMS)
support referential integrity, or the idea of ensuring that
values linking the tables together are valid and correctly
synchronized.
• Structured Query Language (SQL) is the standard language
for accessing the data in the tables.

11-8
 (cont’d)
Relational
database
example

11-9
 Object Databases
• The object database, or object-oriented
database, is based on the premise of object
orientation that all things should be treated as
objects that have both data (attributes) and
processes (behaviors).
• Changes to one object have no effect on other
objects because the attributes and behaviors self-
contained, or encapsulated, within each one.
• This encapsulation allows objects to be reused.
11-
10
(cont’d)
• Object Database Example

11-
11
 Multidimensional Databases
• A multidimensional database is a type of relational
database that is used extensively in data
warehousing.
• Data warehousing is the practice of taking and
storing data in a data warehouse (i.e., a large
database) that supports decision support systems
(DSS).
• Data marts are smaller databases based on data
warehouse data, and support DSS for specific
departments or functional areas of the
organization.
11-
12
 Selecting a Storage Format

Comparing Data Storage Formats

11-
13
Moving from logical ER to
physical ER

11-
14
MOVING FROM LOGICAL TO
PHYSICAL

11-
15
Conceptual, Logical and Physical
• Conceptual, logical and physical model or ERD are
three different ways of modeling data in a domain.

• Business analyst uses conceptual and logical model

for modeling the data required and produced by
system from a business angle

• while database designer refines the early design to

produce the physical model for presenting physical
database structure ready for database construction.
11-
16
 Conceptual Model

• Conceptual ERD models

information gathered
from business
requirements.
• Entities and relationships
modeled in such ERD are
defined around the
business's need.
• The need of satisfying the
database design is not
considered yet.

11-
17
 Logical Model

• Logical ERD also models information gathered from business requirements. It is

more complex than conceptual model in that column types are set. Note that
the setting of column types is optional and if you do that, you should be doing
that to aid business analysis. It has nothing to do with database creation yet.
11-
18
 Physical Model
• Actual design blueprint of a relational database.
• It represents how data should be structured and
related in a specific DBMS so it is important to
consider the convention and restriction of the DBMS
you use when you are designing a physical ERD.
• This means that an accurate use of data type is
needed for entity columns and the use of reserved
words has to be avoided in naming entities and
columns.
• Besides, database designers may also add primary
keys, foreign keys and constraints to the design.
11-
19
Physical Model

11-
20
Metad
ata

11-
21
OPTIMIZING DATA STORAGE

11-
22
 Optimizing data storage
• The data storage format is now
optimized for processing efficiency.
• There are two primary dimensions in
which to optimize a relational
database: for storage efficiency and for
speed of access.

11-
23
Optimizing Storage Efficiency
• The most efficient tables in a relational
database in terms of storage space
have no redundant data and very few
null values.
• Normalization is the best way to
optimize data storage for efficiency.

11-
24
(cont’d)
Optimizing data storage

11-
25
 Optimizing Access Speed
• After having optimized the data model design for data
storage efficiency, the end result is that data are spread out
across a number of tables.
• For a large relational database, it is necessary to optimize
access speed.
• There are several techniques of optimizing access speed:
• Denormalization
• Clustering
• Indexing
• Estimating the size of data for hardware planning

11-
26
 Denormalization
• Denormalization – adding redundancy back
into the design.

11-
27
 (cont’d)
There are four
reasons for de-
normalization.

11-
28
 Clustering
• Clustering – placing records together
physically so that like records are
stored close together.
• Intrafile clustering – Similar records in
the table are stored together.
• Interfile clustering – Combining
records from more that one table that
typically are retrieved together.
11-
29
 Indexing
• An index in data storage is a minitable
(similar to an index of a book) that
contains values from one or more
columns in a table and the location of
the values within the table.
• Indexes require overhead in that they
take up space on the storage.
11-
30
(cont’d)
Example of indexing

11-
31
(cont’d)
• Guidelines for creating indexes

11-
32
Storage Size Estimation

11-
33
 Estimating Storage Size
• Volumetrics – technique of estimating the
amount of data that the hardware will need to
support.
1. Calculate the amount of raw data - all the data
that are stored within the tables of the database.
2. Calculate the overhead requirements based on
the DBMS vendor’s recommendations.
3. Record the number of initial records that will be
loaded into the table, as well as the expected
growth per month.
11-
34
(cont’d)
• Example of calculating volumetrics

11-
35
 SUMMARY
• File data storage formats
- Files are electronic lists of data.
- Five types of files: master, look-up, transaction, audit,
and history.
• Database storage formats
- A database is a collection of groupings of information
- A DBMS is software that creates and manipulates these
databases.
• Selecting a data storage format
- Relational databases support simple data types very
effectively, whereas object databases are best for
complex data.

11-
36
 (cont’d)
• Physical entity relationship diagrams
- Physical ERDs contain references to how data
will be stored in a file or database table, and
metadata are included.
• Optimizing data storage
- There are two primary dimensions in which to
optimize a relational database: for storage
efficiency and for speed of access.
- There are a number of techniques of
optimizing data storage.
11-
37