================================================
Internet Architecture, Network Concepts and
Facts
================================================
Written by Mikkkeee
For
http://blacksun.box.sk/
and
http://www.securitywriters.org
OverView
==========
------------------------------------------------------
1. Short
summary on the history of the Internet (boring yet
vital)
2. Internet
Protocols and Jargon
3. Host
Machines and Host Names
4. Architecture
of the Internet and Packet Switching
5. Synchronous
and Asynchronous Communication
6. Who
is in Charge? (boring yet vital)
7. The
future of the Internet
8. Further
reading
------------------------------------------------------
This text won't make you
an elite hacker or a cracker, but will clear up many misunderstandings you might
have about the Internet and the way it is structured. Okay before you start
reading, i will be summarizing most of this info because many other tutorials
have been written in detail explaining the sections presented. This is just
another way to say, IAM NOT GOING TO WASTE MY TIME TYPING. So i see you have
downloaded this text to learn more about the wonderful tool the world has ever
seen, but do you really understand the juicy info behind the Internet? Well
after reading this you will!
1: Summary on the history of the Internet (boring yet
vital)
Well every place i go i keep hearing ppl
saying things like i was chatting, or i sent him/her an email, or i was checking
his site on the INTERNET. Now when you are asked what is the Internet? you
reply, Well the Internet is like a program i write a www.blahblah.com and i see
stuff and i download it, not hard, probably took days to do, well someone is
wrong, heheheh.
The Internet is a global assemblage of more than 20
million computers and growing in rapid intercommunication. These forms of
communication links are made up of optical fibers, digital cables, satellite
transmissions, and yup you guessed it, telephone lines. The Internet can be
described in three terms, the hardware infrastructure that supports it, the
software that powers it, and the people who populate it.
When the
Internet began, until the early 90s, the only guys who used it weren't normal
ppl, but they were the scientists, academics, science students and the real deal
computer programmers who were pursuing long-distance collaborations and
research. The Internet isn't something that just popped out of nowhere, it has
been around for decades, but was something overlooked by the media until 1990.
Well the origin of the Internet dates back to 1970, when four computers, yes
four, one each at The University of California at Los Angeles, The University of
California at Santa Barbara, The Stanford Research Institute, and The University
of Utah at Salt Lake City were hooked up over phone lines.
From the
first four computers that powered the first Internet, 13 years later 562 boxes
were connected to it in 1983, and by 1993 that number has reached 1.2 million
and in 1996 it was 12 million, and this number remains doubling every 12-14
months. One critical component underlying today's Internet is the software that
supports network communications. In the Internets's beginning dayz, the software
that powered the networks wasn't user friendly, they didn't have fancy crap with
popup menus and fancy Graphical User Interfaces (GUI), they had cryptic programs
which looked like hell for normal people. One reason for this is because the
only people who used the net were scientists who didn't care about user friendly
interfaces. The software they designed was difficult, which hindered the
popularity of the net.
The first tools that these cool guys made that
innovated the Internet were, Telnet, yup the good old telnet, which was made in
1969 and second, the File Transfer Protocol (ftp), which was made in 1971 and is
still in great use today. (i wasn't born yet, heh) Other popular communication
programs have also evolved with the Internet, which are know by the names of:
Usenet Newsgroups, Internet Relay Chat (IRC), Gophers, and Multi-user Dungeons
(MUD's).
2: Internet
Protocols and Jargon (what those 3 letter words mean)
We
computer guys tend to speak in acronyms, small words that have a lot of valuable
info. Acronyms are ubiquitous on the Internet, and all this wonder will be
explained here.
Many Internet acronyms involve a "P" letter. Lets look
at these, we have, SLIP, TCP, UDP, FTP, ICMP, SMTP, HTTP, PPP, IP, and Mikkkeee
(only playing). Okay now you have witnessed the confusion, what does all this
stuff mean. All of them stand for the word "protocol." So when ever you see a
"P" at the end of a computer acronym, there is a big chance its stands for
protocol. Now the tough part is what is this protocol? Well a protocol in
computer science just refers to formalities and conventions observed by
computers during cross computer communication. So its just like this, in order
for boxes to share data, each box must use the same data transfer rules, or we
won't have communication. Just like the old dayz when ppl used the telegraph,
which communicated through the communication protocol called the Morse code.
Then came the Ham radio operators who also had their protocols and now we have
the Internet, which also needs its little protocols.
(Now I am not going
to kill you via boredom by explain the concept of protocols because they are
already explained by the tutorials found at BSRF, but i will sprinkle the
sprinkles on the cake.)
The Internet Protocol (IP) is the cornerstone
for all Internet communications. In fact, you can see if a box is alive (on the
net) by asking for its IP, yah not going far as to doing something lame as
DoSing the box! Now each computer on the Internet has a unique IP number that
operates as its Internet address. When we say that there are 20 million
computers on the Internet, that figure is really based on the number of IP
addresses in use in that year.
As you have begun to browser the net, or
start your mission in becoming a hacker, you will most likely come across some
reference to the Transfer Control Protocol/Internet Protocol(TCP/IP). TCP is
nothing to be feared, it works closely with IP in order to prepare packets of
data before an Internet data transfer. TCP also interprets those same packets at
the receiving end after the transfer.
Additional protocols are also used
to power the net. These are, the File Transfer Protocol (FTP), which moves filz
from one computer to another. The Simple Mail Transfer Protocol (SMTP) designed
in 1982 to support e-mail communications. The HyperText Transfer Protocol
(HTTP), made in 1990 so that web browsers could read documents on the
web.
The Point-to-point protocol (PPP) and the Serial Line Internet
Protocol (SLIP), which generally are terms used if you need to set up access to
the Internet over a telephone line through an Internet Service Provider (ISP).
These protocols are used when a computer wants to communicate over the Internet,
over a serial transmission line, like an analog telephone line, and the user
wants to use a browser such as Internet Explorer. So his isp will offer a ppp or
a slip account to the users who want to view graphics on the net. (Now you might
have noticed that I didn't go in depth but for further information on protocols
check out the tuts at bsrf for further info.)
3: Host Machines and Host
Names
Now to gain good
knowledge on how the net works you will need to know something about host
machines/host names.
Now a computer that has been assigned an IP address
is called a HOST MACHINE. Each IP address is made up of four integers separated
by periods. Lets take an example, my ip on the net yesterday was 150.211.39.128
just like zip codes tell a person where you live, IP addresses reveal not your
geographical info, but your computer network's info. Now the numbers to the
furthermost part to the left represent the large part of the Internet, while the
numbers to the further part of the right represent a specific host machine.
While all these numbers are fine for computer communications, long
strings of numbers are not easy for us to remember so that's why we have
substituted the long strings of numbers into symbolic names. Symbolic names look
like this www.yahoo.com, which is easier to remember. Symbolic host names are
always converted into their IP addresses for the purpose of network
communications. The Domain Name Service (DNS) is responsible for this task of
mapping symbolic names to their numerical equivalents, and symbolic host names
are often called DNS names or addresses. Lets look at an example, DNS names,
follow certain naming conventions that help us remember the names easily. so we
have
" blah.cs.NYU.edu "
| | | |
| | | |---> this segment refers to an educational site.
| | |--------> this part stands for the school, here its NYU University.
| |-------------> this stands for Computer Science Department.
|-----------------> this stands for the name of the machine.
ohh one more thing,
sometimes lamers talk about "HIGH LEVEL DOMAIN NAMES" so they think they are so
smart well they are only referring to the final part of the full domain name.
This means a high level domain name identifies the type of site and where the
host machine resides. Here is a
list:
-------------------------------------------
.com = a
commercial organization
.edu = an educational site
.org = a nonprofit
organization
.net = a network site
.gov = a government agency
.mil = a
military site
These high level domain names can in many cases be
followed by other suffixes, that will identify their geographical location by
country. This list is very small, (ahh again a complete list can be found at
BSRF) blahblah.net.lk = Sri Lanka blahblah.net.il = Isreal blahblah.net.gb =
Great Britain okay that's
enough!
-------------------------------------------
4: Architecture of the Internet and Packet
Switching
Well while you
browse the Internet you might think every computer on it is equal well your
wrong, not all computers are equal. Well to understand what i mean i have to
explain to you how data gets moved across the Internet by PACKET
SWITCHING.
In order to understand packet switching you will have to
understand what packet switching tries to achieve. When the Department of
Defense was paying for the original network research that led to the Internet,
they wanted a strong network that could withstand regional power blackouts or
other problems. So in order to achieve such a robust communication, it was vital
to design a transmission protocol that could readily find new routes if its
destination was dead for some reason. So those guys came up with a cool plan to
design a form of DYNAMIC ROUTING, which will become the standard means of moving
data across the network. Dynamic Routing just means that routes for data are
always selected at the time of transmission, after taking current network
conditions into consideration.
The network designers also felt it was
important to distribute route-selection capabilities throughout the network, so
that no single site was responsible for the entire network. Let me explain this
further, lets say like a network was designed " hierarchically " so that
everything passes through that central routing site, so if anything happens to
that central routing site well HOUSTON WE GOT A PROBLEM!
This is how a hierarchically network would look like:
----------------------------------------------------------------------------
___
|___| / \ / \ Well you can clearly see that the / /\ \ big box is the central routing site O O O O so if anything happens to the big
box then we got a big problem. ----------------------------------------------------------------------------
So since these guys
were very smart they built the Internet in "heterarchically," which really means
instead of having one central routing site, there are a large number of routing
sites distributed all over the network. The hosts that decide how to route
transmissions across the Internet are called ROUTERS. The Internet today is
currently being run by thousands of these hosts called routers.
---------------------------------------------------------------------------------
O-O Okay the art sucks but you see the idea / -O- \ there isn't one box there are many hosts
O-----O so if one dies, others can still work \ O / ---------------------------------------------------------------------------------
So now you figured
out the difference, dynamic heterarchical routing is what makes the Internet
work so good, and the key idea that binds everything together is packet
switching. All of the data moved across the Internet is moved in units called
PACKETS. The traffic on the net is measured by counting the packets transferred
in a period of time.
If this concept is hard for you to understand let
me explain. Lets say you are downloading a file from a server in California to
your box in New York. The software running on the server breaks the huge file
into packets. Lets say your file was 100k it might be broken down info 6 or 7
packets, then each packet is stamped with the IP address of its destination, as
well as the IP address of the originating host. Packets will also be given
instructions, these instructions describe the size of the original transmission
and where each specific packet should be positioned when the original
transmission is put together when it gets to my computer. Now the big deal is
that each packet is sent out onto the Internet to find its way to the
destination address. This means that the 6 or 7 packets don't have to travel
together so they can go from California to other parts of the country then they
find their way to New York. The packets are first given to a router, which looks
at their destination address and decides where to send the packets. Now the
router at that location then does the same thing. So when all the packets come
to New York, they are then assembled into a single file. Each packet is checked
first to see if it has arrived intact. If there was an error in a packet then a
request to resend that packet is sent to the originating host. This form of
sending files is the best way to work on a net, because it minimizes network
load and by sending small packets the destination won't have to wait for the
whole file to be sent if it had errors. Every computer running on the Internet
uses this form of data transfer.
Important
newbie fact
Well if your
reading this you probably were alive in 1996, well what's the big deal you ask,
lets put this concept of packet switching into the real world. On August 7,
1996, 6 million AOL customers were unable to access the Internet because AOL had
recently loaded new software onto its network routers. This software contained
incorrect routing information. As a result, AOL couldn't route packets on or off
its gateway hosts. The 6 million guys waiting soon found out that their aol
service was interrupted and for 19 hours these guys couldn't surf the net. Now
since only AOL had this problem, the rest of the world who are smart enough not
to use aol were checking their email and chatting and visiting sites, thanks to
the architecture of the net not all of us suffered.
5: Bandwith and Asynchronous Communication
Okay, you're probably saying
what's this topic Mike? Its nothing to fear, i'll explain in simple English a
little child can understand.
Online communication takes some time for
normal people to understand or to just get used to. In the world you live in
called off-line world communication is dominated by two way, real time
communication, ie. you and a friend talking. This conversation is being operated
in two directions, with questions/answers being talked about right away. Now in
the computer sphere, a two way communication mediated by shared signals for
beginning/ending conversations is called SYNCHRONOUS communication. Synchronous
communication is like normal human/animal conversation. This form of
communication relies on shared signals for starting and stopping. Examples of
this communication would be like your browser sending a msg to a host and the
host sends you something, a packet or a file. Programs which deal with this
communication are software found on your box, Cu-Cms and RealVideo are some,
which allow internet users with properly equipped hosts to converse with other
internet users through a video channel. Problems arise in this form of
communication, because as both of us know if you don't have the bandwidth, you
will be staring at a video box with images that are jerky and stroboscopic
(images which look like garbage).
When a communication exchange does not
rely on shared signals for starting and stopping, this communication is called
ASYNCHRONOUS communication. Now your probably confused but don't be. This form
of communication seen online is characterized by long waits for responses. This
communication is seen in email messages which take some people forever to answer
back. This doesn't have anything to do with the servers or the bandwidth, but
has to do with the human being on the other side. Email messages are delivered
in a matter of seconds, but there is no guarantee that an intended recipient
will be ready and waiting to reply as soon as your message is received. So the
reason for this is that it pays to think about what your are saying and how your
saying because it will affect the outcome of the asynchronous communication.
6: Who is in
Charge?
Your probably wondering, yeah I always wanted an
answer to that question, so explain to me plz. hehe!
Okay I am not going
to lie to you, questions about Internet management and maintenance are very
boring so bare with me. You might also want to ask who pays for all this? Who
fixes it when it messes up? Who is responsible for keeping everything working?
Who makes the decisions about the future of this wonderful tool? Well I'll
answer everything right here.
As I explained in the previous parts of
this text, the net is composed of thousands of smaller networks that are
independent from one another, remember the heterarchical structure,heh. Now
these smaller networks are managed by corporations, government agencies and
universities with different hardware platforms and different network
technologies. Now you see that I am going back and all these small and large
networks are held together by a wonderful glue called TCP/IP. But I see your
going to ask, okay, cool but what holds it together administratively, like how
can this huge tool have no central authority? Okay let me answer your question,
remember when i told you that the Internet was made by scientists and
programmers, well those guys have always pushed the spirit of this wonderful
tool. So when these guys made up TCP/IP they made it public so that TCP/IP could
be placed on any hardware platform. Now the Department of Defense had a large
part in this deal but like the Department of Defense, the university researchers
also did a lot, and like the rest they also made their finding public. By doing
this anyone who had access to the internet, could access these documents and put
their own influence into the development of the net. Now you might think that
all this public stuff is done with, well not. This way of life is still managed
by a group that oversees the continuing growth of the Internet, this group is
called, INTERNET SOCIETY (ISOC). This organization oversees a number of smaller
groups also pursuing the growth of the net. These groups are:
-
Disaster Assistance Committee
- Internet Operations Forum
- Internet
Architecture Board (IAB)
- Internet Engineering Task Force (IETF)
-
Internet Assigned Number Authority (IANA)
- Internet Engineering Steering
Group (IESG)
- Internet Research Task Force (IRTF)
- k-12 Committee
ISOC has no sanction or governmental
power so it operates as a resource for ppl who want to participate in the growth
of the net.
The Internet's high-speed backbone is a key component of the
net, and many different organizations have said they are responsible for
maintaining the backbone. In the 70's, research parties funded by the Advanced
Research Projects Agency (ARPA) were given access to ARPAnet. In the late 70's,
the National Science Foundation (NSF) joined with ARPA to create a network that
would be more available for computer scientists in the US. NSF then gave birth
to CSnet in 1979. In 1982, a gateway was established between ARPAnet/CSnet,
which demonstrated the viability of a network of networks like a small version
of the Internet. After the joining of ARPAnet/CSnet, the term "INTERNET" was
then adopted.
By 1983, more networks began to do the same thing by
expanding/establishing their own user communities. In San Francisco, FidoNet
connected FidoBBS (Bulletin Board Systems) across the nation so that many users
of local bulletin boards could exchange messages by email/discussion groups. In
the same year, the Department of Defense switched all of its military networks
to the TCP/IP protocol. By then, TCP/IP software was only included in all UNIX
software distros out of the University of California at Berkley. UNIX was the
operating system of choice for computer scientist, and it still is, so by doing
this a huge jump was put in Internet expansion.
BY 1985 the Internet was
further expanded when NSF became a leader in network communication and they got
funding from Congress to join 100 more universities to the Internet. By doing
this NSF then connected five supercomputer sites in a new network called NSFnet.
With all these costs and further spending the Federal Government, soon realized
it couldn't afford to subsidize the Internet. So now it became a private
industry. So the firms which made up the NSFnet backbone (IBM, MCI, MERIT)
created a nonprofit company called Advanced Networks and Services (ANS). Then in
1992, ANS built a new backbone, ANSnet, with 30 times the bandwidth of the old
NSFnet backbone. ANSnet is the current backbone for the Internet and the first
one not maintained by the government but by private companies.
Now since
your probably bored and confused, let me finish off, since no federally
sponsored organization oversees everything, there are some which monitor some
parts. Lets look at Federal Communication Commission (FCC) they don't have any
authority over the net, but they do regulate the billing practices of telephone
companies. This is really cool, cause the FCC does not allow phone companies to
bill ppl for digital communication differently than they do for voice
communication. So the main deal is this, the net is operated by small spheres
and ppl who impose on themselves to censor stuff or to protect ppl from the bad
sides of the net.
7: The
future of the Internet
Well I
am going to let you do some work now. Since you just read the info on the net
how big you think the info on Internet2 is going to be. let me give you a hint,
it is going to be big. At the moment, a new upgraded version of the internet is
in development, and its going to be called Internet 2 or simply I2. This will be
a heaven when its complete so with out further interruption I have made a site
dedicated to Internet 2. The site is good enough and is still up to date. Chk
out the links for further news on it. The url of the site is http://i2.virtualave.net/ ahh man i forgot
to kill the banners, well enjoy!!!
8: Further reading
Okay I have chosen some
books if you want to learn more, cause you know in 3 decades i summarized them
in 7 or 8 pages so i think you might want to read more.
Good books to
read.
Where Wizards Stay up Late, by Katie Hafner and matthew Lyon.
Simon and shuster, New York, NY, 1996. (304 pages)
Being Digital, by
Nicholas Negroponte. Vintage Books, New York, Ny, 1995. (255 pages)
The
Internet 1997 Unleashed, by Jill Ellsworth and Billy Baron (Eds)Sams.net
Publishing, Indianapolis, IN, 1996. (1,269 pages)