UNIT - VI

7 Advances in Databases

Syllabus
Emerging Databases : Active and Deductive Databases, Main Memory Databases, Semantic
Databases.
Complex Data Types : Semi-Structured Data, Features of Semi-Structured Data Models. Nested
Data Types : JSON, XML. Object Orientation : Object-Relational Database System, Table
Inheritance, Object-Relational Mapping. Spatial Data : Geographic Data, Geometric Data.

Contents
7.1 Emerging Databases
7.2 Complex Data Types
7.3 Nested Data Types

7.4 Object Orientation

7.5 Spatial Data
Multiple Choice Questions

(7 - 1)
Database Management Systems 7-2 Advances in Databases

7.1 Emerging Databases

 Data is growing rapidly. Day by day it is becoming complex to handle such huge
amount of data properly.
 In order to use the data efficiently the database systems are supported by modern
tools and techniques. Emerging databases are those databases that are heavily
influenced both by the evolution of software applications, and by advances in
computing hardware and operating system design.
 In this chapter we will get introduced by advances in databases by learning the
concept of active and deductive databases, complex and nested data types, software
technologies such as JSON and XML, object oriented database management systems
and spatial databases.

7.1.1 Active and Deductive Databases

Active Databases
 Active databases are the databases which consists of triggers. The situation and
action rules are embedded in the active databases. The active databases are able to
react automatically to the situations in the database. The trigger is a technique for
specifying certain types of active rules. The commercial databases such as Oracle,
DB2, Microsoft SQLServer allows the use of triggers.

Generalized Model for Active Database

The general model for active database is considered as Event-Condition-Action(ECA)
model. This model has three components –
i) Event : The events are database update operations that are performed explicitly on
the databases.
ii) Condition : The condition determines whether the rule action should be executed.
If the action is not specified then the action will be executed automatically on
occurrence of the event.
iii) Action : The action is usually a sequence of SQL statements. It could be a database
transaction or external program that will be executed on occurrence of condition.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-3 Advances in Databases

CREATE TABLE PersonTab (

pname VARCHAR2(20) Here table named PersonTab is created

);

CREATE OR REPLACE TRIGGER MyTrigger

BEFORE INSERT ON PersonTab

FOR EACH ROW

ENABLE
Code for actual trigger
DECLARE

usr_name VARCHAR2(20);

BEGIN

SELECT user INTO usr_name FROM dual;

DBMS_OUTPUT.PUT_LINE('Inserted a new row by user: '||usr_name);

END;

/
SQL query on execution of which the
trigger gets fired.
INSERT INTO PersonTab VALUES('Sharda');

Deductive Database
 Deductive database is a database system that can make deductions based on rules
and facts stored in the database.
 Deductive databases use the concept of logic programming for specifying the rules
and the facts. Prolog is a popular programming language which is based on the
concept of logic programming.
 There are two types of specifications used in deductive databases –
1) Facts : Facts are specified as the same way the relations are specified in the
Relational Database except it is not necessary to include the attribute [Link]
meaning of an attribute value in a tuple is determined solely by its position in the
tuple.
2) Rules : They specify “virtual relations” that are not actually stored but that can be
formed from the facts by applying deduction mechanisms based on the rule
specifications.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-4 Advances in Databases

For example –
/*Facts*/
man(anand)
man(arun)

woman(anuradha)
woman(jayashree)

parent(anand, parth)
parent(anuradha,parth)

parent(arun,anuradha)
parent(jayashree,anuradha)
/* General Rule */
father(F,C):-man(F),parent(F,C)
mother(M,C):-woman(M),parent(M,C)

Now if we fire a query -?father(X,parth)

Then the answer is X=anand. That means ‘anand is father of parth’.

7.1.2 Main Memory Databases

 Main memory database system is a kind of database system in which data resides
permanently on main physical memory.
 The backup copy of such database is maintained on the disk.
 Access to main memory is much faster than the disk access, hence transactions gets
completed quickly. It is essential to have backup copy of the database because if the
main memory gets failed then the entire database system gets lost.
 Some popularly used main memory database management systems are CSQL,
TimesTen

7.1.3 Semantic Databases

 Semantic database management system is a system that stores the meaning of
information as facts about objects.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-5 Advances in Databases

 Semantic databases represent the data models in which data is arranged in some
logical manner.
 It is a conceptual database system that include semantic information that adds basic
meaning of the data and relationships among the data.
 The semantic database systems allow easy development of application programs
and also it becomes easy to maintain such database system when data is updated.

Features
1. Semantic database systems are exceptionally usable and flexible.
2. They have shorter application design and programming cycle.
3. It provides user control via an intuitive structure of information.
4. It empowers the end-users to pose complex, ad-hoc, decision support queries.
5. These are highly efficient database systems.

Review Questions

1. Explain active and deductive database systems.

2. Write short note on – semantic databases.

7.2 Complex Data Types

7.2.1 Semi-Structured Data

 Semi structured data is a kind of data that can’t be organized in relational
databases. This is a kind of data which does not have structural framework but it
might have some structural properties.
 Semi-structured data is basically a combination of structured and unstructured
data. For example - Facebook that organizes information by User, Friends, Groups,
Marketplace, etc., but the comments and text contained in these categories is
unstructured.
 The semi-structured data does not reside in relational database. By applying some
processes we can store them in relational database.
 XML(eXtensible Markup Language) is widely used to store and exchange semi-
structured data.
 Examples of semi-structured data are -
1. Emails
2. Web pages

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-6 Advances in Databases

3. XML documents
4. Zipped files
5. Binary executables

Advantages
1. The semi-structured data is flexible as schema can be changed accordingly.
2. Data is portable.
3. The heterogeneous information can be stored in semi-structured data.
4. There is no need to express the requirements in pure SQL only.

Disadvantages
1. As there is no fixed schema, storing the data is complex.
2. Querying the database is less efficient as the data is in semi-structured manner.
3. Interpretation of relationship among data is difficult.

Difference between Structured and Unstructured Data

Sr. Structured data Semi-structured data Unstructured data

No.

1. It is having fixed and It is combination of It is not predefined or

organized form of data. structured and organized form of data.
unstructured data.

2. It is schema dependent and It is more flexible than It is the most flexible data.
less flexible. structured data but less
flexible than
unstructured data.

3. Structured query languages The tags and elements are Only textual queries are
are used to access the data used to access the data. possible.
present in the schema

4. Storage requirement for data Storage requirements for Storage requirements for the
is less. the data is significant. data is huge.

5. Examples : Phone numbers, Examples : Server logs, Examples : Emails and

Customer Names,Social Tweets organized by messages, Image files, Open
Security numbers. hashtags, emails sorted ended survey answers.
by the inbox, sent or draft
folders.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-7 Advances in Databases

7.2.2 Features of Semi-Structured Data Models

1. The semi-structured data can not be stored in the form of rows and columns in
databases.
2. The semi structured data does not obey the tabular structure of data models. But it
has some structure.
3. Semi-structured data contains tags and elements which is used to group data and
describe how data is stored.
4. The entities can be grouped together based on their properties.
5. The entities in the same group may or may not have the same attributes(properties).
6. As the semi-structured data does not have well defined structure, it can not be
programmed easily by the traditional programming languages.

Review Question

1. Enlist the features of semi-structured data model.

7.3 Nested Data Types

 Nested data types are structured data types for some common data patterns. Nested
data types support arrays,structs and maps.
 For example – array is a homogeneous collection of elements of same data type
elements. For instance : phone number can be stored as [98-230-11111]
struct employee
{
Name string,
id int,
struct address
{
House-no int,
Street string,
City string
}
}

7.3.1 JSON
 JSON stands for JavaScript object notation.
 Using JSON we can store and retrieve data. This text based open standard format.
 It is extended from JavaScript language.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-8 Advances in Databases

Features of JSON
1. It is text based, lightweight data interchange format.
2. It is language independent.
3. It is easy to read and write.
4. It is easy for machines to parse and generate.
5. It uses the conventions that are familiar to the languages like C, C++, Java,
JavaScript, Perl, Python and so on.

Structure of JSON
JSON is built on two structures :
1. A collection of name/value pairs. In various languages, this is realized as an object,
record, struct, dictionary, hash table, keyed list, or associative array.
2. An ordered list of values. In most languages, this is realized as an array, vector, list,
or sequence.

JSON Object
 JSON object holds the Key value pair.
 Each key is represented as string and value can be of any datatype.
 The key and value are separated by colon.
 Syntax
{ string : value, .......}
 For example
“Age”:38
 Each key value pair is separated by comma.
 The object is written within the { }curly brackets.

1) JSON object containing value of different data types

{
“student”: {
“name”: AAA”,  String value
“roll_no”: 10,  Numeric value
“Indian”: true  Boolean value
}

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7-9 Advances in Databases

2) JSON Nested Object

{
“student”: {
“name”: “AAA”,
“roll_no”: 10,
“address”: {
“Street”: “Shivaji Nagar”,
“City”: “Pune”,
“Pincode”: 411005
}
}

7.3.2 XML
 XML stands for eXtensible Markup Language.
 This scripting language is similar to HTML. That means, this scripting language
contains various tags. But these tags are not predefined tags, in-fact user can define
his own tags.
 Thus HTML is designed for representation of data on the web page whereas the
XML is designed for transport or to store data.

Uses of XML
1. XML is used to display the meta contents i.e. XML describes the content of the
document.
2. XML is useful in exchanging data between the applications.
3. The data can be extracted from database and can be used in more than one
application. Different applications can perform different tasks on this data.

Advantages of XML
1. XML document is human readable and we can edit any XML document in simple
text editors.
2. The XML document is language neutral. That means a Java program can generate
an XML document and this document can be parsed by Perl.
3. Every XML document has a tree structure. Hence complex data can be arranged
systematically and can be understood in simple manner.
4. XML files are independent of an operating system.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 10 Advances in Databases

Goals of XML
Following are the goals of XML -
1. User must be able to define and use his own tag. This allows us to restrict the use of
the set of tags defined by proprietary vendors.
2. Allow user to build his own tag library based on his web requirement.
3. Allow user to define the formatting rules for the user defined tags.
4. XML must support for storage or transport of data.

Features of XML
Following are some features which are most suitable for creating web related
applications.
 XML is EXtesible Markup Language intended for transport or storage of the data.
 The most important feature of XML is that user is able to define and use his own
tag.
 XML contains only data and does not contain any formatting information. Hence
document developers can decide how to display the data.
 XML permits the document writer to create the tags for any type of information.
Due to this virtually any kind of information can be such as mathematical formulae,
chemical structures, or some other kind of data can be described using XML.
 Searching sorting, rendering or manipulating the XML document is possible using
Extended Stylesheet Language (XSL).
 The XML document can be validated using some external tools.
 Some commonly used web browsers like Internet Explorer and Firefox Mozilla
provide support to the tags in XML. Hence XML is not at all vendor specific or
browser specific.

7.3.3 Difference between XML and HTML

Sr. HTML XML
No.

1. HTML stands for Hypertext Markup XML stands for eXtensible Markup
Language. Language.
2. HTML is designed to display data with the XML is used to transport and store data, with
focus on look and feel of data. focus on what data is.
3. HTML is case insensitive. XML is case sensitive.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 11 Advances in Databases

4. HTML has predefined tags. XML has custom tags can be defined and the
tags are invented by the author of the XML
document.
5. As HTML is for displaying the data it is As XML is for carrying the data it is dynamic.
static.
6. It can not preserve white space. It can preserve the white space.

7.3.4 Example of XML

<bank>
<account>
<account-number> S101 </account-number>
<branch-name> Shivaji Nagar </branch-name>
<balance> 5000 </balance>
</account>
<account>
<account-number>C102 </account-number>
<branch-name> Model Colony </branch-name>
<balance> 4000 </balance>
</account>
<customer>
<customer-name> Ram Kumar </customer-name>
<customer-street> Fergusson Road</customer-street>
<customer-city> Pune </customer-city>
</customer>
<customer>
<customer-name> Shiv Prasad </customer-name>
<customer-street> Main Road </customer-street>
<customer-city> Nasik </customer-city>
</customer>
<depositor>
<account-number> S101 </account-number>
<customer-name> RamKumar </customer-name>
</depositor>
<depositor>
<account-number> C102 </account-number>
<customer-name> Shiv Prasad </customer-name>
</depositor>
</bank>

Fig. 7.3.1 XML representation of bank information

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 12 Advances in Databases

7.3.5 Building Blocks of XML Document

Various building blocks of XML are -

1. Elements
The basic entity is element. The elements are used for defining the tags. The elements
typically consist of opening and closing tag. Mostly only one element is used to define a
single tag.

2. Attribute
The attributes are generally used to specify the values of the element. These are
specified within the double quotes.
For example -
<flag type=”True”>
The type attribute of the element flag is having the value True.

3. CDATA
CDATA stands for Character Data. This character data will be parsed by the parser.

4. PCDATA
It stands for Parsed Character Data (i.e. text).

7.3.6 Concept of Namespace

 Sometimes we need to create two different elements by the same name. The xml
document allows us to create different elements which are having the common
name. This technique is known as namespace.
 In some web documents it becomes necessary to have the same name for two
different elements. Here different elements mean the elements which are intended
for different purposes. In such a case support for namespace technique is very much
helpful.
 For example : Consider the following xml document -
[Link]
<File-Description>
<text fname="[Link]">
<describe>It is a text file</describe>
</text>
<text fname="[Link]">
<describe>It is an image file</describe>
</text>
</File-Description>

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 13 Advances in Databases

The above document does not produce any error although the element text is used for
two different attribute values. The output will be,

7.3.7 Document Type Definition (DTD)

 The document type definition is used to define the basic building block of any xml
document.

 Using DTD we can specify the various elements types, attributes and their
relationship with one another.

 Basically DTD is used to specify the set of rules for structuring data in any XML file.

 For example : If we want to put some information about some students in XML file
then, generally we use tag student followed by his/her name, address, standard
and marks. That means we are actually specifying the manner by which the
information should be arranged in the XML file. And for this purpose the
Document Type Definition is used.

 There are two ways by which DTD can be defined.

Internal DTD
Consider following xml document –

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 14 Advances in Databases

XML Document [[Link]]

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE student [
<!ELEMENT student (name,address,std,marks)> Here we are defining the DTD
<!ELEMENT name (#PCDATA)> internally
<!ELEMENT address (#PCDATA)>
<!ELEMENT std (#PCDATA)>
<!ELEMENT marks (#PCDATA)>
]>
<student>
<name>Anand</name>
<address>Pune</address>
<std>Second</std>
<marks>70 percent</marks>
</student>
Output

External DTD
In this type, an external DTD file is created and its name must be specified in the
corresponding XML file. Following XML document illustrates the use of external DTD.

Step 1 : Creation of DTD file [[Link]]

Open some suitable text editor or a notepad. Type following code into it -
<!ELEMENT student (name,address,std,marks)>
<!ELEMENT name (#PCDATA)>
<!ELEMENT address (#PCDATA)>
<!ELEMENT std (#PCDATA)>
<!ELEMENT marks (#PCDATA)>
Now save this file as [Link]

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 15 Advances in Databases

Step 2 : Creation of XML document -

XML Document [[Link]]

Now create a XML document as follows -
<?xml version="1.0"?>
<!DOCTYPE student SYSTEM "[Link]"> The external DTD file is created
<student>
<name>Anand</name>
<address>Pune</address>
<std>Second</std>
<marks>70 percent</marks>
</student>

Step 3 : Using some web browser open the XML document.

Output

Merits of DTD
1. DTDs are used to define the structural components of XML document.
2. These are relatively simple and compact.
3. DTDs can be defined inline and hence can be embedded directly in the XML
document.

Demerits of DTD
1. The DTDs are very basic and hence cannot be much specific for complex
documents.
2. The language that DTD uses is not an XML document. Hence various frameworks
used by XML cannot be supported by the DTDs.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 16 Advances in Databases

3. The DTD cannot define the type of data contained within the XML document.
Hence using DTD we cannot specify whether the element is numeric, or string or of
date type.
4. There are some XML processor which do not understand DTDs.
5. The DTDs are not aware of namespace concept.

Review Questions

1. What is JSON ? Explain it with suitable example.

2. What is XML? Enlist its features.

7.4 Object Orientation

 An object-oriented database (OODBMS) or object database management system
(ODBMS) is a database that is based on object-oriented programming (OOP). The
data is represented and stored in the form of objects.
 OODBMS are also called object databases or object-oriented database management
systems.

Challenges in Designing OODBMS

1. There is no universally agreed data model for OODBMS. Hence no universal
standard is followed while designing the OODBMS based applications.
2. The OODBMS is complex due to handling of complex datatypes, information
hiding, inheritance and other object oriented features.
3. The OODBMS does not provide view mechanism. Therefore the developer can not
have limited view of the physical database system.
4. There is lack of security mechanism in maintaining OODBMS.
5. The OODBMS has scalability and resource usage limitation, including database
server.
6. As amount of data generated and collected explodes, OODBMS is struggling to
keep up.
7. Although there are benefits to decentralize data management and communication
among them using data objects, it presents challenges as well. How will the data be
distributed? What are the best methods of decentralization? There is a major
challenge in designing and managing the database of such systems.
8. In order to handle OODBMS, it requires skilled people who know object oriented
technology thoroughly well.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 17 Advances in Databases

Difference between ODBMS and DBMS

OODBMS DBMS

OODBMS stands for Object Oriented Database DBMS is any Database Management System.
Management System that supports creating The most popular DBMS are relational database
and modelling of data as objects. management systems in which we store
everything as a relation between entities.

Handles larger and complex data than RDBMS. Handles comparatively simpler data.

Data Handling - Stores data as well as methods Data Handling - RDBMS stores only data.
to use it.

OODBMS is object-oriented. RDBMS is table-oriented.

OODBMS uses inheritance and encapsulation Normalization is used to eliminate data

to reduce data redundancy. redundancy in RDBMS.

Examples - Object Database, Objectivity/DB, Examples - MSSQL, MySQL, and Oracle

ObjectStore, Cache, and, ZODB

7.4.1 Object-Relational Database System

 Object relational database system is a database system which is based on object
relation model. It provides the migration path for the users of relational databases
who wish to use object oriented features.
 There are two commonly addressed issues for supporting object relational database
systems -
1. Build object oriented database systems(OODBMS) that natively support object
oriented type system and allows direct access to data from object oriented
programming language using the native type system of the language.
2. Automatically convert data from the native type system of programming language
to relational representation and vice [Link] data conversion is specified as
object relational mapping.

7.4.2 Table Inheritance

 Table inheritance is a feature where there are sub-tables in Relational database
management systems that inherit the schema of parent tables.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 18 Advances in Databases

 For example – Following statements demonstrate how to achieve table inheritance

create table people of Person;  Parent table
create table students of Student  Child table
under people;
create table teachers of Teacher  Child table
under people;
 As a result of above statements, every attribute present in the people table is also
present in the subtables students and teachers. Similarly every tuple present in the
students or teachers become implicitly present in people.
 Using the keyword only in the query we can restrict the access to particular table.
For example – if we want to find the tuples that are present in the people table but
not in its subtables then we should use “only people”

7.4.3 Object-Relational Mapping

 Object oriented relational mapping is an approach in which object oriented
programming language is integrated with databases.
 The object-relational mapping allows a programmer to define mapping between
tuples in database relations and objects in the programming language.
 An object can be retrieved based on the selection condition on its attributes.
 The object oriented programs can be used to create an object, update the objects and
delete the object. Similarly mapping from object to relations is possible by updating,
deleting, and inserting tuples in the database.
 Hibernate is an open source object relational mapping tool for the Java platform.

Advantages
1. It supports object oriented features such as inheritance, polymorphism, association
and so on.
2. Instead of plain data, the object relational mapping return objects. These objects can
be easily accessed, programmed and permanently stored in memory.
3. Object relational mapping systems also provide query languages that allow
programmer to write queries directly on the object model. Such queries are
translated into SQL queries for underlying relational database.

Disadvantages
1. The query language for object relational mapping systems have limited capabilities.
2. There is significant overhead on the object relation system when there occurs bulk
database update.

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 19 Advances in Databases

Review Questions

1. Explain the concept of table inheritance with suitable example.

2. Explain object relational mapping. Give its advantages and disadvantages.

7.5 Spatial Data

 Spatial data means data related to space.
 The special data in the database system allows it to store, index and query the data
based on the basis of special locations.
 Spatial data represents location, size and shape of an object on the earth. For
example - City, location of school, or location of hospital is represented with the
help of spatial data.

7.5.1 Geometric Data

 Various geometric data constructs can be represented in a database in normalized
fashion.
 Following is a list of various geometric constructions and the description on how to
store the information of these geometric constructs in the database systems –
Line : The line segment is represented by co-ordinates of its endpoints.
Curve : Approximate a curve by partitioning it into a storage. Create a list of
vertices in order. Represent each segment as a separate tuple that also carries with it
the identifier of the curve.

Closed polygons :
i) It is represented as list of vertices in order starting vertex is the same as ending
vertex.
ii) Represent boundary edges as separate tuples with identifier.
iii) Use triangulation – Divide polygon into triangles. Use identifier for each triangles.

Line segment {(a1,b1), (a2,b2)}

Triangle {(a1,b1), (a2,b2), (a3,b3)}

TECHNICAL PUBLICATIONS® - an up- thrust for knowledge

Database Management Systems 7 - 20 Advances in Databases

Polygon {(a1,b1), (a2,b2), (a3,b3), (a4,b4), (a5,b5)}

Polygon {(a1,b1), (a2,b2), (a3,b3), ID1} {(a1,b1), (a3,b3),

(a4,b4), ID2} {(a1,b1), (a4,b4), (a5,b5), ID3}

 The 3D shapes are represented in the similar manner as of 2D shapes. But it has
extra z component.
7.5.2 Geographic Data
 Geographic data are spatial in nature. For example – maps and satellite images are
geometric data.
 Maps not only provides the location but along with location it also provides
detailed information about the location.
Applications of geographic data
1) Web based road map services which allows us to use map data for vehicle
navigation.
2) Vehicle navigation systems store information about roads and services for use of
drivers.
3) Global Positioning System (GPS) information broadcasts from GPS satellites to find
the current location of user with some accuracy.
Representation of geographic data
 There are two types of geographic data
1. Raster data : Raster data consists of bit maps or pixel maps in two or more
dimensions. For example – 2D raster image : Satellite image of cloud cover, where
each pixel stores the cloud visibility in a particular area.
2. Vector data : The vector data is constructed from basic geometric objects such as
points, line segments, polygon, triangles and so on. The vector data is always used
to represent the map data. For instance – roads can be considered as two
dimensional(2D) and represented by lines and curves. Rivers may be represented as
complex curves. The regions and lakes can be represented using polygons.
Review Question

1. Explain geographic data in detail.



TECHNICAL PUBLICATIONS® - an up- thrust for knowledge