BABA GHULAM SHAH BADSHAH

UNIVERSITY

BGSBU
Department OF :- Information Technology & Engineering

Name :- Danish Manzoor

Roll No. :- 17-ITE-2022

Course Title :- DBMS

Course Code :- PCC-ITE-421

Assignment :- Ist

Submitted to :- Mr. Rakesh Sir

1|Page
Q.1 Investigate and compare different file organizations such as sequential, indexed, and
hashed file organizations. Provide examples and discuss the advantages and
disadvantages of each.

Sequential File Organization:

➢ In a sequential file organization, records are stored in sequential order based on a key
field or in the order they were added. Each record contains a fixed length of data.

Example: An example could be a text file where each line represents a record, and records are
added or accessed sequentially.

ID | Name | Age

------+------------+-----

001 | John Doe | 30

002 | Jane Smith | 25

003 | Alice Lee | 35

Advantages:

• Simple to implement and understand.

• Suitable for applications where data is accessed sequentially, such as batch processing.

Disadvantages:

• Not efficient for random access; accessing records not at the beginning or end of the
file requires scanning through all preceding records.
• Not suitable for applications requiring frequent updates or random access.

Indexed File Organization:

➢ In an indexed file organization, records are stored in a sequential manner, but an index
is maintained separately to facilitate fast access to records based on a key field.

2|Page
 Index:
ID | Pointer
------+---------
001 | 100
002 | 200
003 | 300

Data:
Address | Name | Age
--------+------------+-----
100 | John Doe | 30
200 | Jane Smith | 25
300 | Alice Lee | 35

Example: A database table where records are stored sequentially on disk, but an index structure
(e.g., B-tree or hash table) is maintained in memory for quick lookup.

Advantages:

• Allows for both sequential and random access.

• Efficient for large datasets where random access is required.

Disadvantages:

• Index maintenance overhead: Updating the index whenever records are added, deleted,
or modified can be resource-intensive.
• Indexes consume additional storage space.

Hashed File Organization:

➢ In a hashed file organization, records are stored in a table-like structure where the
address of each record is determined by applying a hash function to its key field. This
allows for direct access to records based on their keys.

3|Page
Hash Table:

Hash | Data

-------+-----------

0 | (empty)

1 | (empty)

2 | (empty)

3 | Alice Lee, 35

4 | John Doe, 30

5 | (empty)

6 | Jane Smith, 25

Example: A hash table where records are stored based on the result of a hash function applied
to their keys.

Advantages:

• Provides direct access to records based on their keys, leading to fast retrieval.
• Well-suited for applications requiring frequent random access.

Disadvantages:

• Collision handling: When multiple records hash to the same address, collision
resolution techniques (such as chaining or open addressing) are needed, which can
impact performance.

4|Page
 • Hash function design: The efficiency of a hashed file organization relies heavily on the
quality of the hash function chosen.

Q.2 Explain the concepts of primary and secondary indexes in database systems. Describe

how they differ and provide examples of their usage in real-world scenarios.

Primary Index:

➢ A primary index is an index structure that is built on the primary key of a table. The
primary key uniquely identifies each record in the table, and the primary index
organizes the data based on this key. Each entry in the index points directly to the
corresponding record in the table. Here's an example:

Table: Employees

ID (Primary Key) | Name | Department

-----------------+------------+-----------

001 | John Doe | Sales

002 | Jane Smith | Marketing

003 | Alice Lee | HR

Primary Index:

ID | Pointer

------+---------

001 | 0

002 | 1

003 | 2

5|Page
Usage in Real-world Scenarios:

• Database Lookup: Primary indexes are used for efficient retrieval of records based on
their primary key.
• Enforcing Uniqueness: Primary keys ensure the uniqueness of records, and primary
indexes help enforce this constraint efficiently.
• Join Operations: Primary indexes are often used in join operations to efficiently merge
data from multiple tables based on primary key relationships.

Secondary Index:

➢ A secondary index is an index structure that is built on a non-primary key column of a

table. Unlike primary indexes, which are based on the primary key, secondary indexes
can be built on any column that is frequently used in queries for efficient data retrieval.
Each entry in the secondary index points to the location of records based on the indexed
column. Here's an example:

Table: Employees
ID (Primary Key) | Name | Department
-----------------+------------+-----------
001 | John Doe | Sales
002 | Jane Smith | Marketing
003 | Alice Lee | HR

Secondary Index (on Department):

Department | Pointer
----------- + ---------
Sales | 0
Marketing | 1
HR | 2

6|Page
Usage in Real-world Scenarios:

• Query Optimization: Secondary indexes speed up queries that filter or sort

data based on indexed columns, improving query performance.
• Range Queries: Secondary indexes facilitate efficient retrieval of records
within a specified range of values.
• Avoiding Full Table Scans: Secondary indexes help avoid full table scans by
providing direct access to relevant records based on the indexed column.

Differences:

• Key Columns: Primary indexes are based on the primary key column(s) of a
table, while secondary indexes are based on non-primary key columns.
• Uniqueness: Primary indexes enforce the uniqueness constraint, whereas
secondary indexes do not necessarily enforce uniqueness.
• Usage: Primary indexes are typically used for primary key lookups and join
operations, while secondary indexes are used for optimizing query performance
on non-primary key columns.

In summary, primary and secondary indexes serve different purposes in database

systems, with primary indexes primarily used for primary key lookups and secondary
indexes used for optimizing query performance on non-primary key columns. Both
types of indexes play crucial roles in improving database performance and query
efficiency.

Q.3 Compare and contrast various index structures including hash-based indexing,
dynamic hashing techniques, multi-level indexes, and B+ trees. Discuss the efficiency,
scalability, and suitability of each structure for different types of data and query
operations.

7|Page
1. Hash-based Indexing:

➢ Hash-based indexing uses a hash function to map keys to their corresponding storage
locations. The hash function determines where data is stored and retrieved directly.

Efficiency:

• Access Time: Access time is generally constant, O(1), making hash-based indexing
efficient for exact match queries.
• Insertion and Deletion: Efficient, typically O(1) on average.
• Space Efficiency: Can be more space-efficient compared to other structures.

Scalability and Suitability:

• Scalability: Hash-based indexing can struggle with hash collisions as the dataset
grows, leading to performance degradation.
• Suitability: Well-suited for exact match queries on large datasets where keys have a
uniform distribution and collisions are minimal.

2. Dynamic Hashing Techniques:

➢ Dynamic hashing techniques, like extendible hashing and linear hashing, dynamically
adjust the hash table's size to accommodate data growth.

Efficiency:

• Access Time: Similar to hash-based indexing, access time is typically constant, O(1),
for exact match queries.
• Insertion and Deletion: Can be efficient, but may involve occasional restructuring of
the hash table, leading to increased overhead.
• Space Efficiency: Can dynamically adjust to optimize space utilization.

Scalability and Suitability:

• Scalability: Dynamic hashing techniques handle data growth well, as they can
dynamically resize to accommodate more data.
• Suitability: Suitable for datasets that experience frequent insertions and deletions, as
well as for datasets with unpredictable growth patterns.

8|Page
3. Multi-level Indexes:

➢ Multi-level indexes organize data using multiple levels of indexing structures, such as
primary and secondary indexes.

Efficiency:

• Access Time: Access time varies based on the number of levels, typically logarithmic,
O(log n).
• Insertion and Deletion: Can be efficient, but may require updates to multiple levels of
indexes, leading to increased overhead.
• Space Efficiency: Can be space-efficient, particularly when combined with techniques
like sparse indexing.

Scalability and Suitability:

• Scalability: Multi-level indexes can scale well for large datasets, but may suffer from
increased access time as the number of levels grows.
➢ Suitability: Suitable for datasets with complex querying patterns, where different
levels of indexing can optimize various types of queries.

4. B+ Trees:

➢ B+ trees are balanced tree structures commonly used for indexing in databases. They
provide efficient range queries and support for sequential access.

Efficiency:

• Access Time: Access time is typically logarithmic, O(log n), making B+ trees efficient
for range queries and exact match queries.
• Insertion and Deletion: Efficient, typically logarithmic, O(log n), due to balanced tree
properties.
• Space Efficiency: B+ trees can be space-efficient, particularly for large datasets.

9|Page
Scalability and Suitability:

• Scalability: B+ trees scale well for both read and write operations, maintaining
balanced tree properties even with dynamic data.
• Suitability: Suitable for a wide range of datasets and query operations, particularly for
range queries, as well as for datasets with frequent insertions and deletions.

Comparison:

• Hash-based Indexing vs. B+ Trees: Hash-based indexing is efficient for exact match
queries, while B+ trees excel in range queries and provide better support for sequential
access.
• Dynamic Hashing vs. Multi-level Indexes: Dynamic hashing techniques dynamically
adjust to data growth, while multi-level indexes provide flexibility for optimizing
various query patterns.
• Scalability: B+ trees and dynamic hashing techniques are more scalable for dynamic
datasets, while hash-based indexing and multi-level indexes may face scalability issues
with large and dynamic datasets.

10 | P a g e