Module 4 part a

1) List Armstrong Axiom rules.

2) How can we conclude two FDs are equivalent

3)Define Boyce Codd normal form. How does it differ from 3NF

4) What is dependency preservation property for decomposition

Module 4 part b
1)Explain with examples 1NF,2NF, 3NF and BCNF.

2) What is the lossless join property of decomposition? Why is it important?

Module 3 part a
1) Outline concept of assertion in SQL with an example

2) Explain any three differences between Hash indexes and B* tree indexes

3) Describe any three aggregate functions in SQL with examples.

4) Explain a situation where a multi-levelindex would be significantly less effective than a single-level
index, and vice versa.

5) Illustrate the concept of trigger in SQL with an example

6) Compare DDL and DML with the help of an example

7) . Give any three uses of a trigger 3

8) A file has r :20000 STUDENT records of fixed length. Each record has the following fields: NAME
(30 bytes), SSN (9 bytes), ADDRESS (40 bytes) PHONE(9 bytes), BIRTHDATE (8 bytes), GENDER (1
byte), DEPTID (4bytes), CLASSCODE (4 bytes), and PROGID (3 bytes). An additional byte is used as a
deletion marker. The file is stored on the disk with block size 8:512 bytes, Calculate the record size R
in bytes. Calculate the blocking factor bfr and the number of file blocks by assuming an unspanned
organization. Calculate the average time it takes to find a record by doing a linear searc

Module 4 part b
1)With the help of an example explain Single-level indexing and multi-level indexing. Explain the
difference (any four) between single Level indexing and multi-level indexing.

2)Explain any three differences between B Tree and B+ tree indexes.

Differences Between B-Tree and B+ Tree Indexes

B-Trees and B+ Trees are both widely used indexing structures in databases, but they have key
differences in how they store and retrieve data.

1. Structure and Data Storage

Feature B-Tree B+ Tree

Data Stores both keys and data in internal Stores only keys in internal nodes; actual data is
Storage and leaf nodes. stored only in leaf nodes.

Leaf All leaf nodes are linked, forming a linked list for
Not necessarily linked.
Nodes sequential access.

Example: In a B-Tree, a key and its associated data may exist in an internal node. In contrast, in a
B+ Tree, only leaf nodes store the actual records.

2. Search Efficiency

Feature B-Tree B+ Tree

Search Searching can end at any node, as data Searching always continues until a leaf node
Process is present in internal nodes. is reached, where data is stored.

Faster for range queries because all data is

Performance Slower for range queries.
in linked leaf nodes.

B+ Trees are faster for range queries (e.g., finding all records between two values) because leaf
nodes are linked, making traversal efficient.

3. Space Utilization and Height

Feature B-Tree B+ Tree

Height of Taller because internal nodes store Shorter because internal nodes store only
Tree both keys and data. keys, making branching higher.

Space Less space-efficient since data is More space-efficient as data is stored only in
Efficiency stored at multiple levels. leaves.

A B+ Tree has more keys per node, reducing the tree height and making it more balanced and
efficient.

Summary of Key Differences

Feature B-Tree B+ Tree

Data is stored in both internal and leaf

Data Storage Data is stored only in leaf nodes.
nodes.

Search Process Search may stop at any node. Search always goes to leaf nodes.

Faster for range queries due to linked leaf

Range Queries Slower for range queries.
nodes.

Space Less efficient, as data exists in multiple More efficient, as internal nodes only
Utilization levels. store keys.

Conclusion:

• B-Trees are better when frequent insertions and deletions are needed at various levels.

• B+ Trees are better for fast retrieval and range queries, making them widely used in
databases.

Would you like an example with diagrams for better clarity?

Single-Level Indexing vs. Multi-Level Indexing

Indexing is used in databases to optimize the search process by reducing the time required to locate
records. It works similarly to an index in a book, allowing quick access to desired information.

1. Single-Level Indexing

In single-level indexing, there is only one level of index that directly maps search key values to
actual data records.

Example:

Consider a student database with the following records:

Roll No Name Marks

101 John 85

102 Alice 90

103 Bob 78

104 Eve 92

We can create an index table on "Roll No":

Index (Roll No) Block Address

101 001

102 002

103 003

104 004

Here, each index entry maps the Roll No to a specific block address in the database, reducing search
time compared to a linear scan.

Limitations of Single-Level Indexing:

• If the number of records is large, the index table itself becomes too large to fit in memory.

• Searching still requires scanning the index table, which can be slow for very large datasets.

2. Multi-Level Indexing

Multi-level indexing solves the problem of large index tables by introducing multiple layers of
indexes.

Example:

If the index table in single-level indexing becomes too large, we introduce a second-level index:

First-Level Index (Main Index Table)

First Level Index Block Address

101 – 150 01

151 – 200 02

201 – 250 03

Second-Level Index (Block Address 01)

Roll No Block Address

101 001

102 002

103 003

104 004

How Multi-Level Indexing Works:

1. First, search the main index table to find the block that contains the required roll number.

2. Then, search within that specific block in the secondary index to find the exact record.
Advantages of Multi-Level Indexing:

• Reduces the number of comparisons by organizing indexes hierarchically.

• Suitable for very large databases where the single-level index is too big to fit in memory.

• Faster search performance compared to single-level indexing.

Differences Between Single-Level and Multi-Level Indexing

Feature Single-Level Indexing Multi-Level Indexing

Structure Contains only one index table. Has multiple levels of index tables.

Slower for large databases as index Faster as it reduces search time using
Speed
size grows. hierarchical indexing.

Storage Requires more storage as index Uses optimized storage, as higher-level

Requirement grows. indexes are smaller.

Large-scale databases with millions of

Best Suited For Small to medium databases.
records.

Would you like a more detailed breakdown of any specific part?

The Lossless Join Property (or Non-Additive Join Property) of decomposition in database
normalization ensures that when a relation is decomposed into two or more sub-relations, the
original relation can be perfectly reconstructed (i.e., joined back) using a natural join operation
without any loss of information.

Definition

A decomposition of a relation RR into sub-relations R1,R2,...,RnR_1, R_2, ..., R_n is lossless if:

R1⋈R2⋈...⋈Rn=RR_1 \bowtie R_2 \bowtie ... \bowtie R_n = R

This means that when we join all decomposed relations, we get back exactly the original relation
without any spurious (extra) tuples.

Conditions for Lossless Join

A decomposition of RR into R1R_1 and R2R_2 is lossless if:

(R1∩R2)→R1or(R1∩R2)→R2(R_1 \cap R_2) \to R_1 \quad \text{or} \quad (R_1 \cap R_2) \to R_2

That is, the common attributes R1∩R2R_1 \cap R_2 must act as a superkey in at least one of the
decomposed relations.

Example

Lossless Decomposition

Consider a relation:
R(A,B,C)R(A, B, C)

with a functional dependency:

A→BA \to B

If we decompose RR into:

R1(A,B)andR2(A,C)R_1(A, B) \quad \text{and} \quad R_2(A, C)

The common attribute is A, and since A → B, A acts as a key in R1R_1, ensuring that the
decomposition is lossless.

Lossy Decomposition

If we decompose R(A,B,C)R(A, B, C) into:

R1(A,B),R2(B,C)R_1(A, B), \quad R_2(B, C)

Here, the common attribute is B, but since there is no functional dependency like B→AB \to A or
B→CB \to C, the decomposition is lossy. Joining R1R_1 and R2R_2 may introduce extra tuples.

Why is Lossless Decomposition Important?

• Prevents data redundancy while ensuring no information is lost.

• Ensures that queries on the decomposed relations return correct results.

• Helps in normalization (such as 3NF or BCNF) while maintaining data integrity.

Would you like a more detailed example with a step-by-step check for lossless decomposition?

Situations Where Multi-Level Indexing is Less Effective than Single-Level Indexing (and Vice Versa)

Indexing improves database search efficiency, but choosing between single-level and multi-level
indexing depends on factors like data size and query patterns.

1. When Multi-Level Indexing is Less Effective than Single-Level Indexing

Situation: Small Database with Few Records

• Suppose we have a small database with only a few hundred records.

• Example: A company stores employee records with less than 500 entries.

• A single-level index (one index table pointing to data records) is sufficient because the index
itself is small and can fit into memory.

• Using multi-level indexing in this case would increase complexity without providing much
benefit.

Why?
 • Multi-level indexing introduces extra levels that must be traversed, adding overhead
without improving performance for small datasets.

2. When Single-Level Indexing is Less Effective than Multi-Level Indexing

Situation: Very Large Database (Millions of Records)

• Suppose we have a huge e-commerce database storing millions of product listings.

• A single-level index would be too large to fit in memory and require scanning a huge index
table.

• Multi-level indexing optimizes the search by allowing a hierarchical lookup, reducing search
time.

Why?

• Instead of scanning a massive single-level index, a multi-level index allows faster access by
narrowing down the search at each level.

• This significantly improves performance for large-scale applications.

Key Takeaways

Scenario Best Index Type Why?

Small database (few records, Single-Level Multi-level indexing adds unnecessary

<1000) Index complexity.

Large database (millions of Multi-Level Reduces index size in memory and speeds up
records) Index search.

Would you like a real-world analogy to clarify this further?