Threads, Processes, Concurrency & Parallelism

Key Definitions

Processes

• A process is any running program on your computer
• A single application may consist of multiple processes
• Your CPU shares its workload across all running processes
• Examples of processes running simultaneously: Chrome, Slack, VS Code, recording software

Threads

• A thread is a unit of execution within a process
• Every process has at least one thread
• A process can be split into multiple threads for better performance
• The CPU works on individual threads — not directly on processes

Process
├── Thread 1
├── Thread 2
└── Thread 3

CPU Architecture

Physical Cores

• The actual processing units built into the CPU hardware
• The number of physical cores = number of operations that can happen simultaneously
• Common counts: 2, 4, 6, 8, up to 64+ cores

Logical Cores (Hyper-Threading)

• A technology (common in Intel CPUs) where each physical core acts as 2 logical cores
• Each physical core can execute 2 operations at the same time
• Logical cores = physical cores × 2

4 physical cores  →  8 logical cores (with hyper-threading)

How many threads can run at once?

The number of logical cores = the number of threads that can execute in parallel at any moment

How the CPU Manages Threads

• The CPU is always working — if one thread stalls (waiting for a file, network, etc.), it immediately switches to another thread
• Managing which threads run when is called scheduling — handled automatically by the CPU and OS
• As a programmer, you decide how to split your program into threads

Concurrency

One thread executing at a time, but multiple threads exist and take turns.

Thread A: ──────────── (stalls)
Thread B:             ──────────── (stalls)
Thread A:                         ────────────

• Requires only one logical core
• Threads don't run simultaneously — they alternate
• When one thread stalls, the CPU switches to another

Practical example — saving a file while using the UI

Parallelism

Multiple threads executing at the exact same time across multiple logical cores.

Core 1:  Thread A ──────────────────────
Core 2:  Thread B ──────────────────────

• Requires multiple logical cores
• Threads don't wait for each other — they truly run simultaneously
• Example: Thread A prints 1, 2, 3, 4 while Thread B prints 5, 6, 7, 8 at the same time → output could be 1 5 2 6 3 7 4 8
• More powerful than concurrency but more complex to implement correctly (synchronisation required)

Concurrency vs Parallelism

Key Takeaways & Recap