Software Engineering Stack Exchange is a question and answer site for professionals, academics, and students working within the systems development life cycle. It only takes a minute to sign up.
Sign up to join this community
Using python syntax,
threading.Barrier(NUM_THREADS) establishes a control point and all participating threads block until all of the participating “parties” have reached that point. It lets threads start up separately and then pause until they are all ready to proceed.
Can someone explain, a real time scenario(multi-thread), on usage of barrier?
Barriers are used to make multiple threads within the same program do something in a synchronized fashion.
The real-world analog would be the starting gate at a horse race. Ten horses in a race won't arrive at the gate at the same time, but it isn't until all ten of them have entered the gate and are ready to race does the gate open and each horse starts its race. (Starting gates are also called starting barriers because they were barriers rather than gates, which might be where the threading construct got its name.)
Threading's barrier does the same thing. In a system where each thread has a function and all must be ready before any starts operating, the way to enforce that is have the threads initialize and then meet at a barrier. You might see this in the system that runs the engine in your car, where the thread overseeing ignition wouldn't want to operate until it can be sure the threads that handle the throttle and crank position sensors are ready to provide useful data.
An example usage is something like a worker pool where you would want all of your workers to report ready before sending them work. Or something like the subsystems of a game loading up in parallel. Or the map steps in map-reduce...
It's just a tool.
