The scheduler turns the circulation RR is one of The the most simple , the most fair, the most and the most common algorithms of the scheduler is of the type of non-exclusive (preemptive). This algorithm like FCFS, but to any processing, the maximum amount of time a specific CPU is given.

In other words, a small unit time, called a quantum of time (time quantum) , cutting time (time slice) is defined, which is usually between 10 to 100ml H, and each process maximum to the amount of could be the CPU, at the discretion of the government. When processing CPU's at its disposal may occur .

or the explosion of computing the current is less than a quantum time is that in this mode the processing of voluntary CPU release and does not wait for completion of operation I/O can be (like FCFS) or that the explosion of computing is more of a quantum is when in this mode, the timer a interrupt to the operating system and gives the operating system with the replacement text (Context switch) CPU from processing the current and taken it to the bottom of The sends, then from the beginning of the queue ready, The processing of another's direction of run pick :

from this method in systems share time used, until the Times of response to users dialog as appropriate the warranty.

the upper limit of the quantum time its so is the response time good.
far down the cutting time by a factor of two is determined by the one that have the cut much larger than the switching time of the text is, for example, thousands of times.
the other thing is that the amount of cutting time should be slightly larger than the time necessary for a interplay, somehow be, because otherwise every, little need to have several cutting time will be the efficiency of the system due to replacement of text in multiple low.

a good rule of thumb is that the go % blasts computing should be shorter than quantum, when they are in operation for: would cut the time to approximately 100 milliseconds in the fall.