GLOBAL ACADEMY OF TECHNOLOGY

Department of Computer Science & Engineering(AI&ML)

Bengaluru – 560098 | Affiliated to VTU

MODULE 5 – SEMINAR PRESENTATION SCHEDULE

Database Management Systems (DBMS)

Transaction Management & Concurrency Control

Academic Year 2025 – 2026

Semester & Section IV Semester – Section B

Total Students 68 Students

Total Teams 34 Teams (2 Members Each)

Unique Topics 34 (One per Team)

Duration per Presentation 8 – 10 Minutes

Head of Department
Department of CSE(AI&ML), Global Academy of Technology
MASTER PRESENTATION SCHEDULE
Each team has been assigned a unique topic from Module 5. Topics are ordered to cover the syllabus completely and
progressively.

Team Section Presentation Topic Member 1 Member 2

1GA24CI064 JNANIKA S
T1 A Introduction to Transaction Processing 1GA24CI063 JESTAVI C
S

1GA24CI066
1GA24CI065 JOHN
T2 A Transaction States and State Transitions
WESLY S
KEERTHANA
BASAVARAJ

1GA24CI068 LAKSHMAN
Desirable Properties of Transactions: Atomicity & 1GA24CI067 KUSUMA R
T3 A NAYAKA
RAMAKRISHNA
Consistency SHIVAPUJI

Desirable Properties of Transactions: Isolation & 1GA24CI069 MANIKANTA

T4 A M
1GA24CI070 MANOJ M
Durability

1GA24CI071
T5 A System Log and Recovery Concepts MEENAHALLI MATADA 1GA24CI072 MEGHANA V
AMOGHA

1GA24CI073 NAMRATHA 1GA24CI074 NATHIN

T6 A Concurrency Problems in Transaction Processing
K GOWDA K G

T7 B Introduction to Schedules and Serial Schedules 1GA24CI075 NEHA B M 1GA24CI076 NIRANJAN S

1GA24CI078 PARINITHA
T8 B Recoverable and Irrecoverable Schedules 1GA24CI077 NITIN J
CS

1GA24CI079 PARINITHA
T9 B Cascading Rollback and Cascadeless Schedules
R
1GA24CI080 PAVAN K

1GA24CI081
1GA24CI082 PRAGNA K
T10 B Strict Schedules PAVANKUMAR
S
MANJUNATH HEGDE

Characterizing Recoverability: Summary and 1GA24CI083 PRAJNA L 1GA24CI084 PRAKASH

T11 B TANTRY RAJ
Comparison

Recoverable Schedules – Practical Examples & 1GA24CI085 PRANVEER

T12 B PRATAP SINGH
1GA24CI086 RAHUL C
DBMS Implementation

T13 C Introduction to Serializability 1GA24CI087 RAJ KUNIKA 1GA24CI088 RAKSHITHA

1GA24CI090 RISHITH
T14 C Conflict Operations and Conflict Equivalence 1GA24CI089 RASHMI N K
GOWDA G

1GA24CI093 SANJANA M
T15 C Conflict Serializability and Precedence Graphs 1GA24CI091 S SHREYA
K

1GA24CI094 SANTHOSH 1GA24CI095 SHARMILA

T16 C View Serializability
N ASHOK

1GA24CI097
1GA24CI096 SHASHANK
T17 C Testing Serializability – Worked Problems
B
SHEELAVANTHARU
MAHANTESH

Serializability vs. Recoverability – Relationship 1GA24CI099 SHREYA

T18 C 1GA24CI098 SHRAVANI T
BHAT
and Trade-offs

1GA24CI100 SINCHANA
T19 D Introduction to Locking and Basic Lock Types
B TALAWAR
1GA24CI101 SINCHANA J

1GA24CI102 SINCHANA 1GA24CI103 SNEHA

T20 D Two-Phase Locking (2PL) Protocol
MS ARUN

T21 D Strict and Rigorous 2PL; Deadlock in Locking 1GA24CI104 SONIYA 1GA24CI105 SOUMYA
PATIL
Team Section Presentation Topic Member 1 Member 2

Protocols

1GA24CI106 1GA24CI107 SPANDANA

T22 D Lock Conversion and Multiple Granularity Locking
SOWMYASHREE R N

Livelock, Starvation and Practical 2PL 1GA24CI109 TANISHA K

T23 D 1GA24CI108 SUMANTH R
JAYAKUMAR
Implementation

Timestamp Ordering – Basics and Timestamp 1GA24CI110 1GA24CI111

T24 E TEJASHWINI Y THEJASHWINI R
Assignment

T25 E Basic Timestamp Ordering (TO) Protocol 1GA24CI112 USHA R 1GA24CI113 VARSHAN G

Thomas Write Rule and Multiversion Timestamp 1GA24CI114 VARSHITH 1GA24CI115 VASU
T26 E C NANDAN
Ordering

Comparison: Locking vs. Timestamp-Based

T27 E 1GA24CI116 VIKAS G 1GA24CI117 VINAY G
Concurrency Control

1GA24CI118 VINNAVAN 1GA24CI119

T28 F Database Failure Types and Recovery Categories
NEWTON. E VINODKUMAR U

T29 F Shadow Paging Recovery Technique 1GA24CI120 VISHAL V 1GA24CI121 VISHWAS J

Log-Based Recovery: Deferred Update and 1GA24CI122 YASH 1GA25CI407-T JAGAN L.

T30 F UPADHYAY R
Immediate Update

ARIES Recovery Algorithm – Overview and Data 1GA25CI408-T LOKESH L 1GA25CI409-T SANJANA
T31 F K B
Structures

1GA25CI410-T SHUBHA 1GA25CI411-T TEJAS

T32 F ARIES Recovery Algorithm – Three Phases
CS GOWDA N

Checkpointing and Comparison of Recovery 1GA25CI412-T

T33 F VENUGOPAL
1GA24CI128 RISHIKA
Techniques

1GA24CI123
Transaction Rollback, Savepoints and Nested
T34 F 1GA24CI125 PREM B S JANYASHREE
Transactions PREETHAM
SECTION A: TRANSACTION PROCESSING FUNDAMENTALS
Teams 1 – 6 | 6 Unique Topics | 8–10 Minutes Each

TEAM 1 — Introduction to Transaction Processing

Module Group: Section A: Transaction Processing Fundamentals

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI063 JESTAVI C 1GA24CI064 JNANIKA S S

Topic Title Introduction to Transaction Processing

Sub-topic Focus Definition of a Transaction; DBMS transaction model; read/write operations

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ What is a transaction and how it differs from a simple query
▶ Logical unit of database processing – atomicity concept
▶ Read_item(X) and Write_item(X) operations in DBMS
▶ Transaction examples: banking transfer, airline reservation
▶ Why transaction management is critical in multi-user systems

TEAM 2 — Transaction States and State Transitions

Module Group: Section A: Transaction Processing Fundamentals

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI065 JOHN WESLY S 1GA24CI066 KEERTHANA BASAVARAJ

Topic Title Transaction States and State Transitions

Sub-topic Focus Active, Partially Committed, Committed, Failed, and Aborted states

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Five transaction states: Active, Partially Committed, Committed, Failed, Aborted
▶ State transition diagram with all valid transitions
▶ Difference between Partially Committed and Committed states
▶ When a transaction moves to Failed and Aborted states
▶ Restart vs. Kill decision for aborted transactions
 TEAM 3 — Desirable Properties of Transactions: Atomicity & Consistency

Module Group: Section A: Transaction Processing Fundamentals

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

LAKSHMAN RAMAKRISHNA
1GA24CI067 KUSUMA R NAYAKA 1GA24CI068
SHIVAPUJI

Topic Title Desirable Properties of Transactions: Atomicity & Consistency

Sub-topic Focus ACID – Atomicity and Consistency properties in depth

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Atomicity: All-or-nothing principle; commit and rollback
▶ How the recovery subsystem enforces atomicity
▶ Consistency: a transaction must preserve database integrity constraints
▶ Difference between consistency of a transaction and a schedule
▶ Real-world examples illustrating Atomicity and Consistency violations

TEAM 4 — Desirable Properties of Transactions: Isolation & Durability

Module Group: Section A: Transaction Processing Fundamentals

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI069 MANIKANTA M 1GA24CI070 MANOJ M

Topic Title Desirable Properties of Transactions: Isolation & Durability

Sub-topic Focus ACID – Isolation and Durability properties in depth

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Isolation: concurrent transactions must not interfere with each other
▶ Levels of isolation (read uncommitted to serializable) overview
▶ Durability: committed changes survive system failures
▶ Role of the buffer manager and write-ahead log in durability
▶ Real-world examples showing isolation and durability failures
 TEAM 5 — System Log and Recovery Concepts

Module Group: Section A: Transaction Processing Fundamentals

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI071 MEENAHALLI MATADA AMOGHA 1GA24CI072 MEGHANA V

Topic Title System Log and Recovery Concepts

Sub-topic Focus Write-Ahead Logging (WAL); log records; UNDO and REDO operations

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Purpose of the system log in transaction management
▶ Log record types: start, write, commit, abort, checkpoint
▶ Write-Ahead Logging (WAL) protocol and why it is essential
▶ UNDO and REDO operations during recovery
▶ Checkpointing to limit recovery time

TEAM 6 — Concurrency Problems in Transaction Processing

Module Group: Section A: Transaction Processing Fundamentals

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI073 NAMRATHA K 1GA24CI074 NATHIN GOWDA K G

Topic Title Concurrency Problems in Transaction Processing

Sub-topic Focus Lost Update, Dirty Read, Unrepeatable Read, Phantom Read

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Why concurrent execution of transactions can cause problems
▶ Lost Update Problem with an illustrated example
▶ Dirty Read (Temporary Update) problem
▶ Unrepeatable Read (Incorrect Summary) problem
▶ Phantom Read problem and how serializability addresses all four
SECTION B: SCHEDULES BASED ON RECOVERABILITY
Teams 7 – 12 | 6 Unique Topics | 8–10 Minutes Each

TEAM 7 — Introduction to Schedules and Serial Schedules

Module Group: Section B: Schedules Based on Recoverability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI075 NEHA B M 1GA24CI076 NIRANJAN S

Topic Title Introduction to Schedules and Serial Schedules

Sub-topic Focus Schedules, serial vs. non-serial schedules, schedule notation

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Definition of a schedule (history) of n transactions
▶ Serial schedule: transactions execute one after another
▶ Non-serial (concurrent) schedules and why they are preferred
▶ Schedule notation: read/write operations with transaction identifiers
▶ Equivalence of schedules and why it matters

TEAM 8 — Recoverable and Irrecoverable Schedules

Module Group: Section B: Schedules Based on Recoverability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI077 NITIN J 1GA24CI078 PARINITHA C S

Topic Title Recoverable and Irrecoverable Schedules

Sub-topic Focus Recoverability condition; cascading rollback scenarios

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ What makes a schedule recoverable or irrecoverable
▶ Formal recoverability condition: commit order must respect read-from relationship
▶ Example of an irrecoverable schedule with step-by-step analysis
▶ Why irrecoverable schedules are never allowed in DBMS
▶ Practical implications for transaction commit ordering
 TEAM 9 — Cascading Rollback and Cascadeless Schedules

Module Group: Section B: Schedules Based on Recoverability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI079 PARINITHA R 1GA24CI080 PAVAN K

Topic Title Cascading Rollback and Cascadeless Schedules

Sub-topic Focus Cascading abort; conditions for avoiding cascadeless schedules

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ What is cascading rollback and when it occurs
▶ Example illustrating cascade of aborts across transactions
▶ Performance cost of cascading rollbacks
▶ Cascadeless (Avoids-Cascading-Aborts) schedule definition
▶ Condition: a transaction may only read committed values

TEAM 10 — Strict Schedules

Module Group: Section B: Schedules Based on Recoverability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

PAVANKUMAR MANJUNATH
1GA24CI081 1GA24CI082 PRAGNA K S
HEGDE

Topic Title Strict Schedules

Sub-topic Focus Strict schedule definition; comparison with cascadeless schedules

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Strict schedule: no read or write on an item until its last writer commits or aborts
▶ Formal definition and condition for strictness
▶ Strict vs. Cascadeless: which is more restrictive and why
▶ Advantages of strict schedules for recovery
▶ Hierarchy: Serial ⊂ Strict ⊂ Cascadeless ⊂ Recoverable
 TEAM 11 — Characterizing Recoverability: Summary and Comparison

Module Group: Section B: Schedules Based on Recoverability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI083 PRAJNA L TANTRY 1GA24CI084 PRAKASH RAJ

Topic Title Characterizing Recoverability: Summary and Comparison

Sub-topic Focus Full comparison of recoverable, cascadeless, and strict schedules with examples

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Side-by-side comparison of all three recoverability categories
▶ Worked examples classifying a given schedule into the correct category
▶ Venn diagram of the containment hierarchy
▶ Which schedule class is used in practice and why
▶ Trade-off between recoverability and concurrency

TEAM 12 — Recoverable Schedules – Practical Examples & DBMS Implementation

Module Group: Section B: Schedules Based on Recoverability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI085 PRANVEER PRATAP SINGH 1GA24CI086 RAHUL C

Topic Title Recoverable Schedules – Practical Examples & DBMS Implementation

Sub-topic Focus Case studies; how DBMS enforces recoverability

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Step-by-step worked examples: classify each schedule
▶ How DBMS transaction managers track dependencies to ensure recoverability
▶ Commit dependency tracking and deferred commit technique
▶ Impact of isolation levels on recoverability guarantees
▶ Common exam-style problems and solutions
SECTION C: SCHEDULES BASED ON SERIALIZABILITY
Teams 13 – 18 | 6 Unique Topics | 8–10 Minutes Each

TEAM 13 — Introduction to Serializability

Module Group: Section C: Schedules Based on Serializability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI087 RAJ KUNIKA 1GA24CI088 RAKSHITHA

Topic Title Introduction to Serializability

Sub-topic Focus Correctness criterion; serial vs. serializable schedules

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Why serializability is the standard correctness criterion
▶ Serial schedules as gold standard of correctness
▶ Serializable schedule: equivalent to some serial schedule
▶ Result equivalence and its limitations
▶ Overview of conflict equivalence and view equivalence

TEAM 14 — Conflict Operations and Conflict Equivalence

Module Group: Section C: Schedules Based on Serializability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI089 RASHMI N K 1GA24CI090 RISHITH GOWDA G

Topic Title Conflict Operations and Conflict Equivalence

Sub-topic Focus Conflicting operations; conflict-equivalent schedules

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Definition of conflicting operations: same data item, at least one write
▶ Non-conflicting operations and when they can be swapped
▶ Conflict equivalence: two schedules have the same conflict order
▶ Step-by-step conflict equivalence check with examples
▶ Limitations of conflict equivalence vs. view equivalence
 TEAM 15 — Conflict Serializability and Precedence Graphs

Module Group: Section C: Schedules Based on Serializability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI091 S SHREYA 1GA24CI093 SANJANA M K

Topic Title Conflict Serializability and Precedence Graphs

Sub-topic Focus Constructing and interpreting serialization (precedence) graphs

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Conflict serializable schedule: conflict-equivalent to a serial schedule
▶ Precedence (serialization) graph construction rules
▶ Adding directed edges for conflicting operations
▶ Testing conflict serializability: cycle detection in the graph
▶ Worked examples: draw graph, detect cycles, determine serializability

TEAM 16 — View Serializability

Module Group: Section C: Schedules Based on Serializability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI094 SANTHOSH N 1GA24CI095 SHARMILA ASHOK

Topic Title View Serializability

Sub-topic Focus View equivalence conditions; view serializable but not conflict serializable

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Three conditions for view equivalence between two schedules
▶ View serializable: view-equivalent to some serial schedule
▶ Relationship: conflict serializable ⊂ view serializable
▶ Example of a view serializable schedule that is not conflict serializable
▶ Why view serializability is NP-complete to test
 TEAM 17 — Testing Serializability – Worked Problems

Module Group: Section C: Schedules Based on Serializability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI096 SHASHANK B 1GA24CI097 SHEELAVANTHARU MAHANTESH

Topic Title Testing Serializability – Worked Problems

Sub-topic Focus End-to-end serializability analysis on complex schedules

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Systematic approach: list operations, identify conflicts, build graph
▶ Multiple worked examples of conflict serializability testing
▶ Finding the equivalent serial order from a cycle-free graph (topological sort)
▶ Common mistakes in building precedence graphs
▶ Practice problems with solutions for exam preparation

TEAM 18 — Serializability vs. Recoverability – Relationship and Trade-offs

Module Group: Section C: Schedules Based on Serializability

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI098 SHRAVANI T 1GA24CI099 SHREYA BHAT

Topic Title Serializability vs. Recoverability – Relationship and Trade-offs

Sub-topic Focus Combining serializability and recoverability for correct schedules

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Neither serializability nor recoverability implies the other
▶ A correct schedule must be both serializable and recoverable
▶ Venn diagram combining all schedule classes
▶ How concurrency control protocols achieve serializability
▶ How recovery protocols achieve recoverability
SECTION D: TWO-PHASE LOCKING TECHNIQUES
Teams 19 – 23 | 5 Unique Topics | 8–10 Minutes Each

TEAM 19 — Introduction to Locking and Basic Lock Types

Module Group: Section D: Two-Phase Locking Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI100 SINCHANA B TALAWAR 1GA24CI101 SINCHANA J

Topic Title Introduction to Locking and Basic Lock Types

Sub-topic Focus Shared locks, exclusive locks, lock compatibility matrix

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Need for locking to prevent concurrent access problems
▶ Shared (read) lock and Exclusive (write) lock definitions
▶ Lock compatibility matrix: which locks can coexist
▶ Lock request, grant, and release operations
▶ Well-formed transactions: locking rules every transaction must follow

TEAM 20 — Two-Phase Locking (2PL) Protocol

Module Group: Section D: Two-Phase Locking Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI102 SINCHANA M S 1GA24CI103 SNEHA ARUN

Topic Title Two-Phase Locking (2PL) Protocol

Sub-topic Focus Growing phase, shrinking phase; proof of conflict serializability

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Two-Phase Locking rule: acquire all locks before releasing any
▶ Growing (expanding) phase: lock acquisitions only
▶ Shrinking (contracting) phase: lock releases only
▶ Lock point concept and why it guarantees conflict serializability
▶ Proof sketch: 2PL produces conflict-serializable schedules
 TEAM 21 — Strict and Rigorous 2PL; Deadlock in Locking Protocols

Module Group: Section D: Two-Phase Locking Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI104 SONIYA 1GA24CI105 SOUMYA PATIL

Topic Title Strict and Rigorous 2PL; Deadlock in Locking Protocols

Sub-topic Focus Strict 2PL, Rigorous 2PL, deadlock detection and prevention

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Strict 2PL: exclusive locks held until commit/abort
▶ Rigorous 2PL: all locks held until commit/abort
▶ How Strict and Rigorous 2PL prevent cascading rollbacks
▶ Deadlock: circular wait condition among transactions
▶ Deadlock detection (wait-for graph) and prevention strategies

TEAM 22 — Lock Conversion and Multiple Granularity Locking

Module Group: Section D: Two-Phase Locking Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI106 SOWMYASHREE R 1GA24CI107 SPANDANA N

Topic Title Lock Conversion and Multiple Granularity Locking

Sub-topic Focus Lock upgrading/downgrading; intention locks; granularity hierarchy

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Lock conversion: upgrading shared lock to exclusive lock
▶ Lock downgrading: releasing exclusive to shared during shrinking phase
▶ Multiple Granularity Locking: database, table, page, row levels
▶ Intention locks: IS, IX, SIX and their purpose
▶ Multiple Granularity Locking compatibility matrix
 TEAM 23 — Livelock, Starvation and Practical 2PL Implementation

Module Group: Section D: Two-Phase Locking Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI108 SUMANTH R 1GA24CI109 TANISHA K JAYAKUMAR

Topic Title Livelock, Starvation and Practical 2PL Implementation

Sub-topic Focus Livelock vs. deadlock; starvation prevention; 2PL in real DBMS

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Livelock: transactions keep retrying but never proceed
▶ Starvation: a transaction is indefinitely denied a lock
▶ FIFO queuing of lock requests to prevent starvation
▶ How 2PL is implemented in commercial DBMS (e.g., MySQL InnoDB)
▶ Performance overhead of locking and ways to reduce it
SECTION E: TIMESTAMP-BASED CONCURRENCY CONTROL
Teams 24 – 27 | 4 Unique Topics | 8–10 Minutes Each

TEAM 24 — Timestamp Ordering – Basics and Timestamp Assignment

Module Group: Section E: Timestamp-Based Concurrency Control

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI110 TEJASHWINI Y 1GA24CI111 THEJASHWINI R

Topic Title Timestamp Ordering – Basics and Timestamp Assignment

Sub-topic Focus Timestamps; read_TS, write_TS; timestamp ordering rule

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Timestamp: unique identifier assigned to each transaction at start
▶ System clock vs. logical counter for timestamp generation
▶ read_TS(X): timestamp of the youngest transaction that read X
▶ write_TS(X): timestamp of the youngest transaction that wrote X
▶ Goal of timestamp ordering: ensure conflict-serializable execution

TEAM 25 — Basic Timestamp Ordering (TO) Protocol

Module Group: Section E: Timestamp-Based Concurrency Control

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI112 USHA R 1GA24CI113 VARSHAN G

Topic Title Basic Timestamp Ordering (TO) Protocol

Sub-topic Focus Read/write rules; transaction rollback on violation

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Read rule: reject if TS(T) < write_TS(X); otherwise grant and update read_TS(X)
▶ Write rule: reject if TS(T) < read_TS(X) or TS(T) < write_TS(X)
▶ What happens when a transaction is rejected: rollback and restart with new timestamp
▶ Worked example: trace a schedule through TO protocol step by step
▶ Advantage of TO: no deadlock possible (no locking involved)
 TEAM 26 — Thomas Write Rule and Multiversion Timestamp Ordering

Module Group: Section E: Timestamp-Based Concurrency Control

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI114 VARSHITH C 1GA24CI115 VASU NANDAN

Topic Title Thomas Write Rule and Multiversion Timestamp Ordering

Sub-topic Focus Optimizing write rejection; multiversion concurrency control overview

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Thomas Write Rule: ignore an obsolete write instead of aborting
▶ Why Thomas Write Rule is correct and when it applies
▶ Multiversion Concurrency Control (MVCC): keep multiple versions of data
▶ How MVCC allows reads to see a consistent snapshot without blocking
▶ Comparison: Basic TO vs. Thomas Write Rule vs. MVCC

TEAM 27 — Comparison: Locking vs. Timestamp-Based Concurrency Control

Module Group: Section E: Timestamp-Based Concurrency Control

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI116 VIKAS G 1GA24CI117 VINAY G

Topic Title Comparison: Locking vs. Timestamp-Based Concurrency Control

Sub-topic Focus Deadlock, starvation, overhead trade-offs between 2PL and TO

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ 2PL: blocking-based; can deadlock; no cascading rollback with Strict 2PL
▶ Timestamp Ordering: non-blocking; deadlock-free; may cause starvation (repeated restarts)
▶ When TO outperforms 2PL: low conflict, read-heavy workloads
▶ When 2PL outperforms TO: high conflict, write-heavy workloads
▶ Hybrid approaches used in modern DBMS (e.g., snapshot isolation)
SECTION F: RECOVERY SYSTEM TECHNIQUES
Teams 28 – 34 | 7 Unique Topics | 8–10 Minutes Each

TEAM 28 — Database Failure Types and Recovery Categories

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI118 VINNAVAN NEWTON. E 1GA24CI119 VINODKUMAR U

Topic Title Database Failure Types and Recovery Categories

Sub-topic Focus Transaction, system, and media failures; recovery classification

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Transaction failure: logical error or system abort within a transaction
▶ System (crash) failure: power loss, OS crash — volatile memory lost
▶ Media (disk) failure: disk head crash — non-volatile data lost
▶ Corresponding recovery techniques for each failure type
▶ Role of the log, buffer, and disk in recovery

TEAM 29 — Shadow Paging Recovery Technique

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI120 VISHAL V 1GA24CI121 VISHWAS J

Topic Title Shadow Paging Recovery Technique

Sub-topic Focus Page table; current vs. shadow page table; commit and abort in shadow paging

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Shadow paging concept: maintain two page tables (current and shadow)
▶ How updates are made to new pages without touching shadow
▶ Commit protocol: make current page table the new shadow
▶ Abort protocol: simply discard the current page table
▶ Advantages and disadvantages vs. log-based recovery
 TEAM 30 — Log-Based Recovery: Deferred Update and Immediate Update

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI122 YASH UPADHYAY 1GA25CI407-T JAGAN L. R

Topic Title Log-Based Recovery: Deferred Update and Immediate Update

Sub-topic Focus NO-UNDO/REDO vs. UNDO/REDO strategies; log record formats

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Log record format: <T, X, old_value, new_value>
▶ Deferred update (NO-UNDO/REDO): changes written to disk only after commit
▶ Immediate update (UNDO/REDO): changes may be written before commit
▶ When to UNDO and when to REDO during recovery
▶ Worked recovery example for each strategy

TEAM 31 — ARIES Recovery Algorithm – Overview and Data Structures

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA25CI408-T LOKESH L K 1GA25CI409-T SANJANA B

Topic Title ARIES Recovery Algorithm – Overview and Data Structures

Sub-topic Focus ARIES design principles; log, transaction table, dirty page table

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ ARIES: Algorithm for Recovery and Isolation Exploiting Semantics
▶ Three ARIES principles: Write-Ahead Logging, repeating history, logging changes during undo
▶ Log Sequence Number (LSN) and pageLSN
▶ Transaction Table: lastLSN, transaction status
▶ Dirty Page Table: recLSN (recovery LSN) for each dirty page
 TEAM 32 — ARIES Recovery Algorithm – Three Phases

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA25CI410-T SHUBHA C S 1GA25CI411-T TEJAS GOWDA N

Topic Title ARIES Recovery Algorithm – Three Phases

Sub-topic Focus Analysis, Redo, and Undo phases of ARIES with worked example

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Phase 1 – Analysis: scan log from last checkpoint to identify winners and losers
▶ Phase 2 – Redo (Repeating History): redo all logged updates from recLSN
▶ Phase 3 – Undo: rollback all losing transactions using CLRs
▶ Compensation Log Records (CLRs): how ARIES avoids re-undoing on second crash
▶ Worked end-to-end example: simulate a crash and trace all three phases

TEAM 33 — Checkpointing and Comparison of Recovery Techniques

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA25CI412-T VENUGOPAL 1GA24CI128 RISHIKA

Topic Title Checkpointing and Comparison of Recovery Techniques

Sub-topic Focus Fuzzy checkpointing; comparison of shadow paging, deferred update, ARIES

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Purpose of checkpoints: reduce recovery time by limiting log scan
▶ Simple (quiescent) checkpointing: suspend all transactions temporarily
▶ Fuzzy checkpointing: non-blocking checkpoint used in ARIES
▶ Side-by-side comparison: Shadow Paging vs. Deferred Update vs. ARIES
▶ Why ARIES is the industry standard and used in IBM DB2, SQL Server
 TEAM 34 — Transaction Rollback, Savepoints and Nested Transactions

Module Group: Section F: Recovery System Techniques

Member 1 – USN Member 1 – Name Member 2 – USN Member 2 – Name

1GA24CI125 PREM B S 1GA24CI123 JANYASHREE PREETHAM

Topic Title Transaction Rollback, Savepoints and Nested Transactions

Sub-topic Focus Rollback to savepoints; partial rollback; nested transaction recovery

Duration 8 – 10 Minutes (Both members must present)

Key Points to Cover in the Presentation:

▶ Full rollback vs. partial rollback using savepoints
▶ How savepoints are recorded in the log and used during undo
▶ Nested transactions: structure, commit/abort propagation
▶ Compensation during nested transaction rollback
▶ Real-world use of savepoints in SQL (SAVEPOINT, ROLLBACK TO SAVEPOINT)
PRESENTATION GUIDELINES
All teams must follow the guidelines below to ensure a professional, complete, and well-evaluated seminar
presentation.
1. Duration & Participation
Each presentation must be 8–10 minutes long. All members are required to present. Presentations shorter than 8
minutes or longer than 10 minutes will incur a penalty during evaluation.

2. Slide Count & Structure

Prepare 9–12 slides: Title Slide, Team Details, Agenda, 5–7 Content Slides (covering all key points), a
Diagrams/Example slide, and a Conclusion & References slide.

3. Content Coverage
Every key point listed on your individual assignment sheet must be addressed. Missing a key point will directly affect your
internal marks.

4. Diagrams & Examples

Include at least one original diagram (state diagram, precedence graph, lock table, ARIES log trace, etc.) relevant to your
topic. Use illustrative numerical or real-world examples.

5. Slide Design
Use a clean, consistent font (minimum 20pt for body text, 28pt for headings). Avoid reading directly from slides. Slides
must be visible from the back of the classroom.

6. Q&A Session
A 2-minute Q&A follows each presentation. Faculty or peers may ask questions. All members must be prepared to
answer.

7. File Submission
Submit your presentation file (PDF or PPTX) to the faculty email / LMS portal at least 24 hours before your scheduled
slot. Late submissions will not be accepted.

8. Dress Code & Conduct

Smart formal attire is mandatory. Speak clearly, maintain eye contact, and avoid excessive use of filler words. Do not
read off the slides.

9. Evaluation Criteria
Marks will be awarded on: Content Accuracy (40%), Presentation Clarity (25%), Slide Quality (15%), Diagrams &
Examples (10%), Q&A Performance (10%).

Evaluation Rubric Summary

Criterion Weightage What Evaluators Look For

All key points covered

Content Accuracy & Depth 40%
correctly

Clear speech, logical flow,

Presentation Clarity 25%
confidence

Slide Quality & Design 15% Readable, structured,

Criterion Weightage What Evaluators Look For

professional

Original, accurate, well-

Diagrams & Examples 10%
explained

Correct and confident

Q&A Performance 10%
responses

Department of Computer Science & Engineering | Global Academy of Technology, Bengaluru

DBMS Module 5 Seminar Schedule | III Semester – Section B | Academic Year 2024–25