(Synchronization is the act of coordiniating two of more program to achieve some goal - note that the goal may benefit only one program at the expense of another...)
Well, if there is one CPU in the computer, there is no way that the computer can run multiple programs at the same time.
But often, the program run by the users of a (fast) computer system are not very "CPU intensive" (requires the CPU to perform many instructions).
Some often run programs are editors (like vi, emacs, etc), that spend most of the time waiting for user's input 9type a key)
In these and many other cases, the CPU can be used to run multiple programs "at the same time" but executing instructions from different programs for very short intervals in rapid succession.
- The figure above depicts how the CPU runs 3 different programs in multi-programming
- It will execute instructions from program 1 for a few milli seconds.
- The the CPU switches to program 2 and execute instructions from program 2 for a few milli seconds.
- and so on...
- To the users of these 3 programs, it gives the appearance that all their programs are running at the same time.
- Multi-programming (or multi-tasking) is common in modern operating systems... even Microsoft Windows has multi-programming since 1995, when they introduced Windows95....
The computer system will incorporate a timer device that periodically generates an interrupt signal (say once every 10 milli second - the period is programmable).
The system is depicted in the following picture:
- The figure shows that their are 3 programs current "active" (ready to be run if it is given the control of the CPU)
- Program P1 is currently running (i.e., the CPU is fetching and executing instructions from program P1)
(1) Acknowledge the interrupt
(2) save the context in the CPU into
the PCB at the head of the Ready Queue
(3) remove the PCB at the head of the Ready Queue
and insert it at the tail of the Ready Queue
(4) restore the context using the PCB at the
head of the Ready Queue
The Interrupt Base Register will contain the starting address of the
above interrupt handling routine.
So the above interrupt handling routine will be run when a timer interrupt is received.
The result is that program P1 (the running program) is suspended and its PCB put to the tail of the Ready Queue.
The PCB of program P2 will become the new head of the Ready Queue and its PCB data will be used to restore the context (registers) in the CPU.
The result is that program P3 will run after the interrupt handling routine finishes execution.
That state is depicted in the following figure:
The result is that each program will be run for a short amount of time - the timer that sends out interrupt signals will "pace" the speed that the CPU is being switched between the various programs.