Introduction to Concurrency

Concurrency and parallelism are terms often used interchangeably, but they describe different approaches to handling multiple tasks. Concurrency refers to the ability of a program to manage multiple tasks at the same time, but not necessarily executing them simultaneously. Instead, tasks may take turns using the same resources, such as a single CPU core, by switching between them quickly. This is particularly useful for programs that spend time waiting for input/output (I/O) operations, like reading from files or network connections.

Parallelism, in contrast, means actually running multiple tasks at the same exact moment. This is possible when you have more than one CPU core available, allowing tasks to be executed truly in parallel. Parallelism is especially important for CPU-bound tasks, which are limited by the speed of the processor, such as performing complex calculations.

Understanding the difference between CPU-bound and I/O-bound tasks helps you choose the right approach. CPU-bound tasks are limited by the speed of computation, while I/O-bound tasks spend most of their time waiting for external operations to complete, like disk access or network responses. In Python, concurrency is often implemented using threads, which are lightweight and can help with I/O-bound tasks. For CPU-bound tasks, multiprocessing, which uses separate processes, is more effective because threads in Python are limited by the Global Interpreter Lock (GIL).

As you saw in the video, choosing between concurrency and parallelism depends on the nature of your program's workload. If your tasks are waiting for I/O, concurrency with threads can improve efficiency. If your tasks need a lot of computation, parallelism with multiple processes will help you get the most out of your hardware.