Explain how implementing systematic store processes ensures transparent records, physical
loss prevention, and absolute audit readiness..

"A poorly managed storehouse acts as a leaky bucket for corporate capital." Comment on this
statement by outlining the essential operational functions of stores management from receipt
to issue.

 Differentiate clearly between Buffer Stock (Safety Stock) and the Reorder Level (ROL).
Discuss how supply lead time variations affect the determination of the reorder point.

Compare and contrast ABC Analysis (Value-based) and VED Analysis (Criticality-based).
How can a store manager integrate these two distinct approaches into an ABC-VED Matrix
for optimal resource allocation?

) Log-Based Recovery and Checkpointing in Data-Intensive Systems

 Query Optimization Techniques for Analytical Queries

 NoSQL Databases for Data Science (Explain one type with use cases)

Star Schema and Snowflake Schema in Data Warehousing

Centralized vs Distributed Databases in Large-Scale Analytics

Explain the Two-Phase Locking (2PL) protocol. Discuss how it ensures consistency in multi-user
analytical environments. Can deadlocks still occur? Explain with a suitable example.
Time T1 T2
t1 READ(X)
t2 X = X + 50
t3 READ(X)
t4 X = X * 1.2
t5 WRITE(X)
t6 WRITE(X)

Compute the final value of X (assume initial X = 100).
(ii) Determine whether the schedule is serializable.
(iii) Identify the concurrency issue involved and explain its impact on analytics accuracy.

Consider the following schedule involving two transactions T1 and T2 operating on a dataset table
storing aggregate metrics:

Explain the ACID properties of transactions with a data science pipeline example, such as feature
store updates or model versioning.

Perform the following using SQL:
 Create tables with appropriate primary and foreign keys
 Insert sample data (minimum 5 researchers, 6 publications)
 List publications by a given researcher
 Find researchers who have not authored any publication
 Find the publication with the highest number of authors
 List venues that have more than two publication

 Consider the following relations used in a research publication analytics system:
 Researcher(ResearcherID, Name, Email, Affiliation)
 Publication(PubID, Title, Year, Venue)
 Authorship(ResearcherID, PubID, AuthorOrder)

 Decompose the above relation into 2NF and 3NF, explaining each step clearly.

 Populate the relation with 8–10 sample records and highlight potential data redundancy and
anomalies.

Consider the relation:
Dataset(DatasetID, DatasetName, Source, CollectionDate, Domain, OwnerName, OwnerEmail,
UpdateFrequency)
 Identify the primary key

 List meaningful functional dependencies

 Explain the importance of indexes in analytical databases. Differentiate between primary index,
secondary index, and clustering index with suitable examples.
Question 2: Data Normalization, Dependencies, and SQL for Analytics

 Convert the ER diagram designed in part (c) into normalized relations up to 3NF, clearly indicating
primary and foreign keys.