Thread (computing): Difference between revisions

Revision as of 12:34, 21 March 2002

Revision as of 18:58, 5 April 2002

Line 2:

software development environments support what are called "threads",

or "lightweight" threads of execution.

They are similar to the idea of a "process", in that both represent

a single sequence of instructions executed monolithically, in parallel

with other processes, either by [[time slicing]] or [[multiprocessing]].

This allows a program to split itself into two or more simultaneously

running tasks.

Line 18 ⟶ 21:

The [[Java programming language]], for example, is designed to

facilitate multi-threaded programs.

An advantage of a multi-threaded program is that it can operate faster on machines

that have multiple CPUs, or across a cluster of machines.

This is because the threaded nature of the algorithms allow true simultaneous and

independent processing.

In such a case, the programmer needs to be careful to avoid [[deadlock]].

In order for data to be correctly manipulated, threads will often need to [[rendezvous]]

in time in order to process the data in the correct order.

Threads may also require an [[atomic]] operation known as a [[semaphore]] in order to

prevent data from being simultaneously modified, or read while in the process of being

modified.