Process vs thread; user and kernel threads

Process and thread

A process is a program in execution, with its own address space, resources and process control block. A thread is a lightweight unit of execution within a process; it has its own program counter, registers and stack but shares code, data and files with other threads of the same process.

User-level threads (ULT)

Threads are created and managed by a user-space library; the kernel knows only the process.

Advantages:

• Fast creation and switching: no kernel trap, only a library call.
• Can run on an OS that does not support threads.
• Each process can use its own scheduling algorithm.
• Scale well to a very large number of threads (no kernel table space).

Disadvantage: a blocking system call by one thread blocks the whole process, and threads cannot run on several CPUs at once.

Kernel-level threads (KLT)

Threads are managed by the kernel, which keeps a thread table.

Advantages:

• The kernel can schedule threads of one process on different CPUs at once (true parallelism).
• If one thread blocks, the kernel can run another thread of the same process.
• The kernel can give more CPU time to processes with many threads and can handle thread-level scheduling and signals.

Disadvantage: slower to create and switch (system calls and mode switches), and uses kernel memory.

Hybrid (many-to-many) models, as in Solaris, combine both.