The operating system’s task scheduler allocates execution time to multiple tasks. By quickly switching between executing tasks, it creates the impression that tasks execute sequentially.

In a computer with a single CPU, multithreading is achieved through time-sharing or time-slicing. The operating system allocates time slices to each thread, allowing them to execute in a seemingly concurrent manner. This is known as time-sharing or time multiplexing.

Here’s a brief explanation:

Time-sharing: The CPU time is divided into small time slices, and each thread is given a portion of that time to execute. The operating system switches between threads rapidly, giving the illusion of simultaneous execution.
Context switching: The operating system maintains the context of each thread, which includes the values of CPU registers, program counter, and other necessary information. When it’s time to switch to another thread, the operating system saves the current thread’s context and restores the saved context of the next thread to be executed.
Thread scheduler: The thread scheduler is responsible for determining the order in which threads are executed. It decides which thread should run next based on priority, time-sharing policies, or other scheduling algorithms.

It’s important to note that true parallel execution is not achieved in a single-CPU system. Instead, the CPU rapidly switches between different threads, giving the appearance of parallelism. This is known as time-sliced or preemptive multitasking.