Linux with Operating System Concepts

virtual memory
and is stored in a 
special area called the 
swap space
. Second, for permanence, we store everything on hard 
disk in user storage space. We differentiate between main memory (cache, DRAM) and 
secondary storage even though both are used to store our program code and data. We will 

---

28
◾
Linux with Operating System Concepts
discuss virtual memory in more detail in Chapters 8 and 10 and the file system in Chapters 
3 and 10.
This organization of memory is known as the 
memory hierarchy
. The idea is that the 
CPU, when it wants something, will try the top level of the hierarchy first (on-chip cache) 
and only if it fails to find what it wants will it resort to moving down the hierarchy. Moving 
down the hierarchy results in slower access but a greater likelihood that what it is look-
ing for will be found because each lower level consists of a storage type that has a greater 
capacity.
If, for instance, the item sought is found in DRAM, then it and its neighboring memory 
locations are copied up the hierarchy. The reason to bring its neighbors is the hopes that 
what the CPU is looking for now is next to memory locations storing items that the CPU 
will look for in the near future. However, in copying items up the hierarchy, other items 
will have to be discarded. For instance, copying some new items into on-chip cache will 
require discarding items from that cache because of its limited size. So the process of mov-
ing items around in memory is something of a guessing game: can we remove items that 
we do not need in favor of items we will need soon?
As mentioned above, computers have some form of memory but not necessarily all of 
these forms. It depends on the type of computer. Handheld devices will often forego hard 
disk storage in favor of using DRAM and in some cases, handheld devices also forego 
DRAM and use only flash memory.
Another component in our computer is the I/O system. We refer to this as a single 
system although in fact it is made up of numerous devices and interfaces (connections). 
Collectively, the devices in the I/O system give the user the ability to interact with the 
computer. The devices will include input devices like a keyboard, mouse or other pointing 
device, touch screen and microphone and output devices like a monitor (possibly a touch 
screen), printer, and speakers. Additionally, our secondary storage devices are part of the 
I/O system as, aside from storing files, they are used for input (when we open a file) and 
output (when we save a file).
So we can think of the computer as being divided into three general categories: the pro-
cessor, memory, and I/O/storage. We can also differentiate the levels of memory and storage 
into a hierarchy of devices. At the top of the hierarchy are the fastest but also most expensive 
forms of memory and so we limit the amount because of their expense. As you descend the 
hierarchy, your computer generally has more and more because the cost per storage unit is 
greatly reduced. This hierarchy consists of from top to bottom: registers, on-chip cache, off-
chip cache, DRAM, swap space, hard disk user space, network storage devices, removable 
storage devices (e.g., flash drive, optical disk drive, and magnetic tape drive).
1.8.3 Software and Users
There are two additional components to any computer system, although these are not hard-
ware. These are software and people (users). The computer itself is a general purpose pro-
cessing machine. It is set up to execute any type of program. But without the program, the 
computer has nothing to do. The hardware by itself has no motivation or goal. So instead, 
humans run programs.

---

Introduction to Linux
◾
29
A program is a step-by-step description of how to solve a problem. The description must 
be in a language that the computer can understand. The only language that a computer is 
capable of executing is that of its native machine language. However, programmers often 
write their program code in other languages and use program translation software to con-
vert their program into machine language. An example program written in C is shown in 
Figure 1.18.
Without a user, there is no one or nothing to tell the computer which program to run and 
at what time. Nor, is there anything to provide input or view the output. It is the user that 
controls the computer. More correctly, it is the user who uses the computer. This is because 
in many cases, the user makes requests of the computer. The computer does not fulfill every 
request as the user must have proper access rights for the request to be carried out.
The computer system then consists of a series of layers. At the top layer is the user who 
uses the computer to accomplish some task(s). The user’s interactions cause applications 
software to execute. Both the user’s interactions and the applications software must access 
computer hardware. The operating system acts as an intermediary between the user/appli-
cation and hardware. The operating system is itself layered with a shell for interaction, 
utilities and services, the kernel, device drivers and ROM BIOS (or firmware), and the 
hardware at the bottom layer (see Figure 1.2).
1.8.4 Types of Computers
Computers can vary in size, power, and expense. At the large end, we have supercomputers 
which consist of dozens, hundreds, or thousands of processors combined by some internal 
network. These computers, which can cost millions of dollars, are the fastest computers 
on the planet. They are often used to handle computationally intensive tasks like weather 
forecasting, modeling earthquakes, operations related to genetics and the human genome, 
and simulating the spread of viruses to name just a few of their applications.
At the other end of the spectrum of computers are handheld devices like smart phones. 
A smart phone is primarily designed to be a telephone but can run a number of other 
#include 
<
stdio.h 
>
int main(){
int a, b, c;
printf(“Enter three numbers: “);
scanf(“%d %d %d”, &a, &b, &c);
if(a 
>
b&&b 
>
c) 
printf(“%d is the greatest\n”, a);
else if(b 
>
a&&b 
>
c)
printf(“%d is the greatest\n”, b);
else
printf(“%d is the greatest\n”, c);
return 0;
}
FIGURE 1.18 
Simple example C program.

---

30
◾
Linux with Operating System Concepts
applications (apps). Typically, the smart phone stores the apps in a small internal memory. 
The processor of the smart phone is one that contains technology of perhaps 10–15 years 
earlier. Because of these limitations, smart phones often can only handle one or perhaps 
two applications at a time. Additionally, to permanently save information recorded on a 
smart phone, you will have to transfer it to another computer. Slightly more capable than 
the smart phone are tablet devices.
In between the two extremes of handheld devices and supercomputers are the personal 
computers: laptops and desktops. This type of computer will contain several storage devices 
including an optical disk drive and an internal hard disk. You can also connect external 
disk drives to the device and for desktop units, additional internal devices such as a second 
hard disk or a floppy disk drive.
More sophisticated than the desktop computer is the server. The idea behind the server 
is that it is a networked computer that services requests of other networked comput-
ers. The server might provide file storage for a few dozen computers or it might handle 
HTTP requests over the Internet (a web server) or it may send and receive email mes-
sages. Depending upon the number of expected users, a server may have similar comput-
ing power and main memory as a desktop computer, or it may require faster processing 
and more memory.
For instance, a file server which is utilized by no more than 10 users may not require 
a superior processor. On the other hand, some types of servers are expected to perform a 
good deal of processing. Web servers for large websites for instance must execute server 
side code. A web site like Google would use high-end servers which contain much more 
powerful processors than what you will find in a desktop computer. What is certain though 
is that a server will utilize a greater amount of disk storage than a desktop computer and it 
is likely that a server might have dozens of hard disk drives.
There are also computers that fall between the supercomputer and the server: main-
frame and minicomputers. These computers are more historically interesting as they 
were the only platform of computer available in the first few decades of computer usage. 
The mainframe and minicomputer category consists of computers expected to handle 
the processing duties of dozens or hundreds of users at a time. With the inexpensive 
availability of personal computers though, mainframe and minicomputer usage is far 
less common.
Table 1.2 illustrates some of the characteristics of each of these types of computers. It 
should be noted that these values tend to change rapidly in the computing field. The values 
shown in Table 1.2 are estimates from 2013. The breakdown of computer types as shown 
in Table 1.2 are not precise. You might notice for instance that high-end mainframe com-
puters and low-end supercomputers are the same. We might think of this grouping as one 
class of computer which ranges at one extreme from computers worth $100,000 to those 
that are worth tens of millions of dollars. We tend to see the greater expense as we add 
more processors (from dozens to thousands) and as we increase the amount of DRAM and 
storage. Similarly, the laptop and desktop computers run together. Laptops can have just 
as great an amount of DRAM, processor cores, and storage space as a desktop computer. 
About the only two aspects that are clearly different between the two types of computers 

---

Introduction to Linux
◾
31
is the laptop is built to be portable and so is lightweight and compact, and the desktop, 
because it is larger, can have a greater number of devices inside (additional disk drive units) 
or connected to it (e.g., external keyboard and speakers).
1.9 THIS TEXTBOOK
Linux (and Unix) textbooks generally fall into two categories. There are users guides that 
are less textbook oriented and more “how to” texts. These are fine if you already know 
something about Linux and want to become more of an expert, or alternatively want to 
learn enough to use the operating system. What they fail to do is teach you underlying con-
cepts; the “why” instead of the “how.” The other category of book is true operating system 
concept books that might spotlight Linux or Unix with one or a few chapters dedicated to 
these operating systems.
The book you hold in your hands is somewhat unique in that it combines both styles of 
textbooks. This book does teach you the “how” of Linux but also emphasizes the “what” 
and “why” of the operating system.
Additionally, Linux (and Unix) textbooks tend to focus on either teaching you how to use 
the operating system or teaching you how to administer the operating system. This textbook 
TABLE 1.2 
Characteristics of Computers (as of October 2013)