⭐ CONCURRENCY CONTROL (Detailed Explanation)

Concurrency Control is a fundamental concept in Database Management Systems (DBMS) and is even more critical in Distributed Databases.
When multiple transactions access the same data simultaneously, concurrency control ensures that the database remains:

✔ Correct
✔ Consistent
✔ Free from conflicts
✔ Serializable

Even when many transactions run together.

⭐ WHAT IS CONCURRENCY CONTROL?

Concurrency Control is the process that manages simultaneous execution of transactions in such a way that:

The ACID properties of transactions are preserved
Database remains consistent
No incorrect updates occur due to interference between transactions

It ensures serializability, meaning the result of parallel execution is the same as some serial execution.

⭐ WHY IS CONCURRENCY CONTROL NEEDED?

When multiple transactions run without proper control, problems arise:

✔ 1. Lost Updates

Two transactions update the same data, but one update overwrites the other.

✔ 2. Dirty Reads

A transaction reads uncommitted data of another transaction.

✔ 3. Non-Repeatable Reads

Data is changed between two reads of the same transaction.

✔ 4. Phantom Reads

Rows added/deleted by other transactions appear/disappear while a transaction is running.

✔ 5. Inconsistent Analysis

When aggregate values are read while other transactions modify data.

Concurrency Control prevents all these anomalies.

⭐ GOALS OF CONCURRENCY CONTROL

✔ Maintain database consistency
✔ Ensure isolation between transactions
✔ Guarantee serializability (correctness)
✔ Manage locks and avoid deadlocks
✔ Maximize transaction throughput
✔ Avoid unnecessary delays

⭐ TYPES OF CONCURRENCY CONTROL TECHNIQUES

Concurrency control techniques are broadly divided into:

Lock-Based Protocols
Timestamp-Based Protocols
Optimistic Concurrency Control
Multi-Version Concurrency Control (MVCC)
Validation-Based Protocols
Deadlock Handling Techniques

Let’s discuss each in detail.

⭐ 1. LOCK-BASED CONCURRENCY CONTROL

A lock is a mechanism that restricts access to a data item.

Two main types:

✔ A. Shared Lock (S-lock / Read Lock)

Allows read-only access
Multiple shared locks allowed

✔ B. Exclusive Lock (X-lock / Write Lock)

Allows read + write
Only one X-lock allowed
No other lock can coexist

⭐ TWO-PHASE LOCKING PROTOCOL (2PL)

(VERY IMPORTANT for exams)

2PL ensures serializability using two phases:

✔ Phase 1: Growing Phase

Transaction obtains all required locks
No locks are released

✔ Phase 2: Shrinking Phase

After releasing the first lock, no new lock can be obtained
Only unlock operations allowed

⭐ Types of 2PL

✔ a) Basic 2PL

Follows two phases strictly.

✔ b) Strict 2PL

Locks are held until transaction commits
Prevents cascading aborts

✔ c) Rigorous 2PL

All locks released only at commit
Strongest form of 2PL

⭐ Advantages

✔ Ensures serializability
✔ Widely implemented

⭐ Disadvantages

✘ Deadlocks possible
✘ Blocking/waiting overhead

⭐ 2. TIMESTAMP-BASED CONCURRENCY CONTROL

Each transaction gets a timestamp when it starts.

✔ Rules:

Older transactions get priority
Database ensures that conflicting operations follow timestamp order

✔ Types:

Basic Timestamp Ordering (TO)
Strict Timestamp Ordering

✔ Advantages:

✔ No deadlocks
✔ Simple to implement

✔ Disadvantages:

✘ More rollbacks
✘ Requires timestamp management

⭐ 3. OPTIMISTIC CONCURRENCY CONTROL (OCC)

Assumes that conflicts are rare.

✔ Phases:

Read Phase – Transaction executes freely.
Validation Phase – Check for conflicts.
Write Phase – If validation succeeds, commit; else rollback.

✔ Advantages:

✔ High performance in low-conflict environments
✔ No locking → no deadlocks

✔ Disadvantages:

✘ Rollbacks increase during high contention

Used in distributed & mobile databases.

⭐ 4. MULTI-VERSION CONCURRENCY CONTROL (MVCC)

(Used by PostgreSQL, Oracle, MySQL InnoDB)

Maintains multiple versions of data items:

✔ Readers get a snapshot (no waiting)
✔ Writers create new versions

✔ Benefits:

✔ High read performance
✔ No read locks required
✔ Non-blocking reads

✔ Drawbacks:

✘ More storage
✘ Complex garbage collection for old versions

⭐ 5. DEADLOCK HANDLING

Concurrency control protocols must manage deadlocks.

✔ Deadlock Prevention

Wait-die protocol
Wound-wait protocol

✔ Deadlock Detection

Wait-for graph
Cycle detection

✔ Deadlock Recovery

Abort a transaction
Rollback and restart

⭐ 6. DISTRIBUTED CONCURRENCY CONTROL (Important)

In distributed systems:

Locks can be maintained globally
Distributed timestamp ordering
Distributed deadlock detection is more complex

Techniques:

✔ Distributed 2PL
✔ Distributed Timestamp Ordering
✔ Token-based methods
✔ Distributed Wait-for graph

⭐ SERIALIZABILITY IN CONCURRENCY CONTROL

Two types:

✔ Conflict Serializability

Schedules can be transformed to a serial schedule by swapping non-conflicting operations.

✔ View Serializability

Weaker but more flexible; ensures final write/read dependencies match serial execution.

Concurrency control aims to achieve serializable schedules.

⭐ ADVANTAGES OF CONCURRENCY Control

✔ Improved system performance
✔ Allows multi-user environment
✔ Prevents anomalies
✔ Ensures correctness
✔ Manages resource usage efficiently

⭐ DISADVANTAGES

✘ Overhead of lock management
✘ Possibility of deadlocks
✘ Rollbacks in optimistic methods
✘ More complex with distributed systems

⭐ Perfect 5–6 Mark Short Answer

Concurrency control ensures that multiple transactions executing simultaneously do not interfere with each other and that the database remains consistent. It prevents problems such as lost updates, dirty reads, and inconsistent states. Concurrency control techniques include lock-based protocols (2PL, strict 2PL), timestamp ordering, optimistic concurrency control, and multiversion concurrency control (MVCC). These techniques ensure isolation and serializability and handle issues like deadlocks and rollbacks.