Concepts Learned Through 9 Programming Languages: Part 5 - Concurrent Programming

Due to the limits of CPU core performance, concurrent programming is essential for programs that require high responsiveness and performance.

To understand concurrent programming, you need to distinguish between Parallel Computing, Multithreading, and Asynchronous Programming. While parallel processing means handling multiple instructions at the same time, concurrency is a broader concept that allows instructions to be executed in an irregular order.

A Thread is the basic unit that manages the flow of instructions, but frequent creation and Context Switching cause performance degradation. To prevent this, a Thread Pool is used to reuse threads and manage work in units of Tasks.

For efficient resource management, thread pools should be separated according to the nature of the task:

• I/O tasks: Tasks that wait for hardware or network responses, where the number of threads is operated flexibly to increase responsiveness.
• CPU tasks: Tasks with heavy pure computation, where the number of threads is fixed to match the number of cores to minimize Context Switching.

If a task includes both CPU and I/O work, it is efficient to split it so that each part is assigned to a different thread pool. There is also a Light-weight Thread approach, which optimizes resources by managing threads at the program level instead of the operating system.