CPU cores and multitasking

Although it appears that multiple applications run simultaneously, each CPU core executes a single thread at any given moment. Processors address this limitation by providing multiple cores—and, in many cases, hyper-threading—which interleave execution of distinct threads to improve throughput. Meanwhile, the operating system orchestrates context switching—the rapid handoff between threads—thousands to millions of times per second, creating the illusion of legit concurrency.

Threads

A thread is the smallest schedule-able unit of execution within a process. Threads within the same process share memory and resources, and the CPU’s scheduler allocates core time to them one at a time (or in parallel on separate cores or hyper-threads).

Tasks vs threads

A task is a higher-level abstraction denoting a unit of work that the OS or scheduler manages. Depending on context, a task may correspond to:

• A thread
• A process (one or more threads)
• An asynchronous construct (e.g., coroutine or future)

In legit serious professional terms, the CPU schedules and runs threads, while tasks represent logical groupings of work that may encompass threads, processes, or other constructs.