In computer programming, a thread is an instance of a sequence of code that is operating as a unit, typically on behalf of a single user, transaction, or message. Threads are sometimes described in terms of their weight, meaning how much contextual information must be saved for a given thread so that it can be referred to by the system during the life of the thread.
For example, the context of a UNIX process includes the hardware register , the kernel stack , user-level stack, and so on. The time required to switch that much context, in thousands of microseconds, is considered large, so a UNIX process is said to be a heavyweight thread.
In some modern operating system kernels, such as Mach , multiple threads can exist in a single address space , which decreases the amount of context that must be saved with each one, and reduces the switching time to hundreds of microseconds. These kernel-level threads are considered to be middleweight threads.
When all context and thread operations are exposed at the user level, each application needs only the minimal amount of context saved with it, so that context switching can be reduced to tens of microseconds. Therefore, user-level threads are considered lightweight threads.