Memory Protection

When several process are running concurrently it is necessary to protect the memory used by one process from unauthorized access by another.The protection mechanism which must be built into the memory addressing hardware are described in detail in the next;for the present we take their existence for granted.I guess the info provided is useful.Thank you for reading.For further clarification or any doubt you can contact me via mail.

Interrupt mechanism ----Part 2

This location will be used as the start of a piece of program known as an interrupt routine,or interrupt handler whose purpose is to determine the source of the interrupt and respond to it in an appropriate maner.We will describe the interrupt handler in section and will discuss the various forms it can take according to the precise nature of the interrupt mechanism available.

Interrupt mechanism ----Part 1

It was mentioned in chapter 2 that in order for I/o activities to be overlapped with central processing it mustbe possible to interrupt the running process when a pheripheral transfer is complete.We therefore demand that our compuer should provide an interrupt mechanism which at least saves the value of the program counter for the interrupted process and transfers control to a fixed locatin in memory.

Small size

Space used to hold the operating system, whether in memory or on backing store, is wasted as far as productive computing is concerned. Furthermore, a large system is liable to be more prone to error , and take longer to write , than a small one .Friends I hope that you are enjoying with this post and gained knowledge about the computing software working.

Assignment of process priorities

The order in which processes are run is determined by the order of the processor queue or by the order in which the dispatcher selects processes from the queue.We suggested in chapter 4 that to minimize dispatching overheads the dispatcher should always pick the first eligible process in the queue;this implies that the order of queue will be the determining factors.The scheduler is responsible for assigning priorities to processes so that when they become runnable they are linked into the appropriate place in the queue.Algorithm for deteminingsuitable priorities will be discussed in the next post.

Maintainability

It should be possible to maintain an operating system enhancing it or correcting errors without employing an army of systems programmers. This implies that the system should be modules and that it should be well documented. So I hope you are very happy with reading of this post and also you will become familiar with operating systems.

Sharing of information

In some system it is desirable that the user should be able to share information.For example the users of a job-shop system may wish to use each other’s programs or data,and in a transaction processing system many separate process may use the same data base. In the job-shop case it may even be desirable for theinstalltion to provide a set of library programs such as editors,compilers or scientific subroutines,which are commonly available to its users.If information shared in this way then it must be stored on-line over long period of time.

Reliability

Ideally an operating system should be completely error-free, and be able to handle all contingencies. In practice this is never the case, but the system should be designed so that errors in one part of it have a minimal effect on other parts. So reliability is very important for the operating system for error free.

On-line storage

For certain applications concerned with te information retrival it is necessary to store large amount of data in such a way that they are accessible.Even in those areas such as job-shop systems where a large on-line data base is not essential,it is great convenience to the user to be able to store program or data within the computing system itself rather than on some external medium such as punched cards.

Desirable features

The need for efficiency has already been alluded to. Unfortunately , it is difficult to light on one single criterion by which the efficiency of an operating system can be judged by various possible criteria like mean time between jobs, unused central processor time, turn-round time for batch jobs ,response time ,resource utilization. Not all these criteria can be satisfied simultaneously. We shall have more to say in later chapters about design decisions which effectively give more weight to some than to others.

Process allocation

Both wait and signal may alter the status of process,the former by making it unrunnable and the later by doing the opposite.An exit must therefore be mad to the dispatcher for a decision on which process status has changed the dispatcher will resume the current process since it will be the first non-runnable process in the processor queue.

Nondeterminacy

An operating system must be determinate in the sense that the same program run today or tomorrow with the same data should procedure the same results. On the other hand it is indeterminate in that it must respond to events which will occur in an unpredictable order. These events are such things as resource requests, runtime errors in programs and interrupts from peripheral devices. Because of the enormous number of contingence which can arise it is clearly unreasonable to expect to write the operating system to cater for them all individually. Instead the system must be written so as to handle any sequence of events.