How does operating system manage program execution

When a process finishes its CPU processing after executing for the allotted time period, this is called scheduling. Priority Based scheduling Non-Preemptive : In this scheduling, processes are scheduled according to their priorities, i. If priorities of two processes match, then schedule according to arrival time. Device Management: The Operating System communicates with the hardware and the attached devices and maintains a balance between them and the CPU.

In order to optimize the CPU time, the operating system employs two techniques — Buffering and Spooling. Buffering: In this technique, input and output data is temporarily stored in Input Buffer and Output Buffer. Once the signal for input or output is sent to or from the CPU respectively, the operating system through the device controller moves the data from the input device to the input buffer and for the output device to the output buffer. In the case of input, if the buffer is full, the operating system sends a signal to the program which processes the data stored in the buffer.

When the buffer becomes empty, the program informs the operating system which reloads the buffer and the input operation continues. When there are various users on a network sharing the same resource then it can be a possibility that more than one user might give it a command at the same point in time.

So, the operating system temporarily stores the data of every user on the hard disk of the computer to which the resource is attached. The individual user need not wait for the execution process to be completed. Instead, the operating system sends the data from the hard disk to the resource one by one.

When a program needs to be executed it is loaded onto the main memory till the execution is completed. Thereafter that memory space is freed and is available for other programs.

The common memory management techniques used by the operating system are Partitioning and Virtual Memory. Partitioning: The total memory is divided into various partitions of the same size or different sizes. This is a multi-step process using several programs. The programs and the files that each create are:. An editor is used to create source files. The source file is written in a programming language, e.

This is very similar to creating a file with a word processor. The main difference is that an editor is based on plain characters, while a word processor formats the text. You see all the characters that you type in an editor, but a word processor includes many hidden formatting codes that define things like paragraphs, bold font, etc.

A compiler or assembler is used to create object files. The compiler or assembler translates the programming language in a source file into the bit patterns that can be used by a CPU machine language. The source file contents remains unchanged. A linker is used to create an executable file.

Most programs are made up of several object files. These are programs that have already been written and compiled.

Computers can store files on the disk secondary storage , for long-term storage purpose. Each of these media has its own properties like speed, capacity, data transfer rate and data access methods.

A file system is normally organized into directories for easy navigation and usage. These directories may contain files and other directions. In case of distributed systems which are a collection of processors that do not share memory, peripheral devices, or a clock, the operating system manages communications between all the processes.

Multiple processes communicate with one another through communication lines in the network. The OS handles routing and connection strategies, and the problems of contention and security. Two ways: First and foremost the answer is "system calls".

Robert Fraser Robert Fraser So, 1. No, it still calls fopen. Somewhere in fopen is a "syscall" instruction. The syscall instruction changes the processor into "kernel mode", which removes all sorts of protections and allows the system to actually access the hardware.

Your program runs in a protected mode and can't access the hardware at all. Where can I find them? I just want to have a glance at the different system calls of different OS for same function say "Opening a file". I'm googling but couldn't find what I'm exactly looking for. For linux: kernel. RobertFraser, Re "parameters to them are OS-specific", But surely there's an easy way to convert to-and-fro between them? Flavius Flavius Christian Hayter Christian Hayter And so why not code to the hardware directly?

Then it would work across OS? Do you want to reboot instead of alt-tab? Or at least go back to the DOS era where a running program basically owned the whole computer. Also, a freestanding program would also need its own drivers for every possible hardware. There are some good examples of how the choice of high-level language, instruction set architecture, and executable file format might affect the ability to run a binary on any given system: Assembly languages must code for a specific ISA.

Matt Joiner Matt Joiner k gold badges silver badges bronze badges. At a high level this is correct. However, you couldn't "write the API yourself" since the OS prevents you from accessing the hardware or page table directly. So at some level you'd still need to make OS-specific syscalls.

An analogy: Say you hire a butler from another country. Re "because you'd rather not buy a new star-trek-translator", so the translator refers to?

Also I think its good to expand this analogy with what happens when you have different ISA too? Pacerier The translator would be the cross-compiler toolchain targeting their ISA.

The point is even though the translator produces x86 or whatever machine language, you need to phrase your instructions differently because kernels have their own interfaces. This means linking to different libraries and using whatever binary format they use. Sign up or log in Sign up using Google. Sign up using Facebook.

Sign up using Email and Password. Post as a guest Name. Email Required, but never shown. The Overflow Blog. Podcast Making Agile work for data science. Stack Gives Back Featured on Meta.