Thinking Concurrently

Concurrent processing means multiple tasks are in progress at the same time. In OCR H446, you need to understand the benefits and challenges of concurrency, including race conditions, deadlock, and shared resource management. Thinking concurrently means identifying which parts of a problem can be solved simultaneously.

---

What Is Concurrent Processing?

Term	Definition
Concurrent	Multiple tasks are in progress during the same time period (may not be executing at exactly the same instant)
Parallel	Multiple tasks are executing at exactly the same instant (requires multiple processors/cores)
Sequential	Tasks execute one after another

Important distinction: Concurrency is about managing multiple tasks; parallelism is about executing them simultaneously. A single-core processor can handle concurrency (by switching between tasks) but not true parallelism.

Sequential:     Task A ------>  Task B ------>  Task C ------>
Concurrent:     Task A -->  Task B -->  Task A -->  Task C -->  Task B -->
Parallel:       Task A ------>
                Task B ------>
                Task C ------>

---

Benefits of Concurrent Processing

Benefit	Explanation	Example
Faster overall completion	Multiple tasks progress simultaneously	A web server handles thousands of requests at once instead of one at a time
Better resource utilisation	While one task waits (e.g., for I/O), another can use the CPU	While waiting for a file to load, the program processes user input
Improved responsiveness	The user interface remains responsive while background tasks run	A word processor saves a document while the user continues typing
Scalability	Adding more processors/cores increases throughput	A multi-core server handles more traffic

---

Identifying Tasks for Concurrency

Not all tasks can run concurrently. The key question is: are the tasks independent?

Scenario	Can be concurrent?	Why?
Downloading 3 files from the internet	Yes	Each download is independent
Computing A + B, then (A + B) * C	No	The multiplication depends on the addition result
Rendering different sections of an image	Yes	Each section is independent
Reading from a file, then writing to the same file	No (must be sequential)	Writing before reading is complete would corrupt data
Searching different parts of a database	Yes	Each search is independent

Key Term: Tasks are independent if neither requires the output of the other. Independent tasks can safely run concurrently.

---

Challenges of Concurrent Processing

Race Conditions

A race condition occurs when the outcome of a program depends on the timing or order of concurrent operations, and that order is not guaranteed.

Example — Bank account:

Two processes try to withdraw from the same account (balance = 100):

Process A: Read balance (100)
Process B: Read balance (100)
Process A: Withdraw 80, write balance (100 - 80 = 20)
Process B: Withdraw 60, write balance (100 - 60 = 40)

Final balance: 40. But 80 + 60 = 140 was withdrawn from an account with only 100. This is a race condition — Process B read the balance before Process A's write was complete.

Correct result: Process A should have completed fully before Process B started, or a locking mechanism should have prevented simultaneous access.

---

Deadlock

Deadlock occurs when two or more processes are each waiting for the other to release a resource, so none of them can proceed.

Conditions for deadlock (all four must be present):

Condition	Description
Mutual exclusion	At least one resource is non-shareable (only one process can use it at a time)
Hold and wait	A process holds one resource while waiting for another
No pre-emption	Resources cannot be forcibly taken from a process
Circular wait	Process A waits for Process B, which waits for Process A

Classic example — Dining Philosophers: