The operating system has to keep track of everything that happens on the computer that it controls. She is alone, and she has to solve a huge number of tasks. The processor is far from the only device that the operating system takes care of. Another of its missions is memory management. She needs to make sure that all processes run smoothly. Why are we talking about processes and not about applications? The answer to this question, our readers found in the article on the principles of the distribution of the operating system of the most precious resource, processor time.
Managing the operating system of computer memory consists of two main tasks:
- Each process must have enough memory to execute it. The memory space occupied by him should not be accessible to another process and he should not wedge himself into the space allotted to another process.
- The system has different types of memory. And they should be used in such a way that each process works with maximum efficiency.
The solution to the first problem is to set operating system restrictions for certain types of software and individual applications.
As an example, let us imagine an incredibly weak computer with 1 megabyte RAM, which we will take as 1000 kilobytes to avoid complicated calculations that will not help us to understand the essence of the issue in question. During the boot process, the operating system of this imaginary computer "bites off" as much memory as it needs. The operating system has priority over other programs. Imagine that the operating system needs only 300 kilobytes of memory to boot.
Imagine RAM as a bath, at the bottom of which the operating system plunged and began to load the various drivers needed to manage the computer's hardware. Drivers needed by the computer in question need 200 kilobytes of RAM. After everything is loaded, without which the computer will not work correctly, the application processes still have 500 kilobytes of RAM.
When applications start to load into memory, each of them occupies a certain number of blocks in RAM, the size of which is assigned by the operating system. If the block size is 2 kilobytes, each process occupies a chunk of memory that is a multiple of two kilobytes. That is, some times 2 kilobytes. Applications are loaded into these fixed-size blocks. Each block begins and ends with a word-limiter, length from 4 to 8 bytes. Blocks and limiters eliminate the situation in which one application could overlap in memory of another. Indeed, in order for such an overlap to occur, it is sometimes sufficient to err by one or two bits. But what to do when the 500-kilobyte memory space allocated for applications is full?
Most modern computers are designed so that you can increase their memory by putting an additional bar in the corresponding slot of the motherboard. For example, if the computer’s RAM is one gigabyte, it can be increased to two. This is the best, but rather expensive and not always supported by the computer configuration, solution. Computer specialists know that most of the application information stored in RAM is not used at any given time.
The processor has access to only one memory area at a time. Thus, most of the available RAM is not used. The hard disk space is much cheaper than RAM, so moving information from RAM to disk can significantly (and most importantly - for free) “expand” RAM. This solution is called virtual memory technology, and we have already written about it.
A hard disk is one of the many types of memory that the operating system manages. Its space is inexpensive, it can store large amounts of information. But the hard disk is a slow repository of information, occupying a very low level in the hierarchy of computer memory.
The operating system balances the needs of multiple processes. To do this, it uses different types of memory, moving data between them in a way that the current situation requires. Data is moved in blocks (otherwise called pages).
To be continued…
Based on computer.howstuffworks.com
The article is based on materials .
Comments
Post a Comment