A variant of round robin scheduling is called selfish round
robin scheduling. In selfish round robin, there is a maximum limit
on the number of processes that can be placed in the round-robin
queue (including the process being executed by the CPU). After that
maximum is reached, newly entering processes are placed on a
holding queue. Processes in the holding queue do not get any time
slice of the CPU. When a process in the round-robin queue completes
and leaves the system, the oldest process in the holding queue is
allowed to enter the round-robin queue