TCP/IP
Home NT WorkStation 4.0 NT Server 4.0 Networking Essentials TCP/IP Need Money FindAJob Nt Server in the Enterprise MCSE,MCP, Certification training resources

 

Home

 

 

 

 

TCP/IP

TechRepublic, THE site for IT professionals

      TCP/IP is the most widely used protocol in computer networks today. This is mainly due to the popularity of the internet. It is my personal belief that soon this Protocol suite will replace all other protocols. unless something better is designed and implemented. For these reasons it is very important to have a good understanding of this area, not just for your certifications but for your career. I originally wrote this paper for a class, so not all the information is just relvant to NT. It encompasses TCP/IP in general.

 

History and Fundamentals of TCP/IP

Introduction

The following pages will outline the history of TCP/IP (Transmission Control Protocol / Internet Protocol) and some of the fundamentals of the protocol. TCP/IP is the most widely used protocol today; this is mainly due to the popularity of the Internet. Because of this, it is very important for the information technology professional to deeply understand this technology.

History

     The idea for TCP /IP (as well as all current transmission protocols) was first thought of by Paul Baran in 1962. In a report to the US Airforce, he described "a robust, efficient, store- and- forward data network." Donald Davies also produced a similar idea for the Postal Service in the U.K. Mr. Davies was the first to coin the term "packets" in which the data would be transmitted. The two men surmised that "packet switching networks could be designed so that all components operated independently, eliminating single point of failure problems." Because of these two men’s work, the modern Internet became possible.

The US Department of Defense started ARPANET (the predecessor to the Internet) in 1968; it first consisted of a server at UCLA and four other nodes at different universities. By 1971, it had grown to encompass the continental United States and had grown into Europe by 1973. The first protocol used on ARPANET was NCP (Network Control Protocol), but it soon became apparent that NCP could not handle the growing network traffic. In 1974, a new protocol was envisioned: TCP / IP. TCP and IP were originally designed to function alone, but as it was soon discovered, the best functionality was to use them together. Thus the protocol suite was invented. The modern TCP /IP protocol suite we use today was created in September 1981. In 1983, the DOD mandated that all of their computers would use TCP /IP for all communications.

 

In 1983, the ARPANET was split into two networks: one was still named ARPANET and the other was named MILNET, which was used by the Military to communicate. Also in 1983, TCP / IP experienced a huge increase in popularity as the UNIX operating system was released with support built in for the protocol. As ARPANET evolved into the modern Internet, TCP/ IP became the universally used protocol for these communications.

 

Fundamentals

Most people think of TCP / IP as just one protocol, but it is, in fact, a suite of protocols and applications that enable a computer to communicate to other computers in a network. It does not matter what operating system the computers are using; as long as each system supports TCP /IP, the computers can communicate and share information. TCP / IP is a broad subject and the following areas of the protocol are not at all exhaustive to all aspects of the suite.

 

IP Addressing

Every host (computer) on the network needs an IP address; TCP/IP has a unique way of identifying a computer on the network. It uses a unique number in the format xxx.xxx.xxx.xxx to identify for which computer the information is intended. Each section of the number or octet tells the system where on the network the computer is located. The first octet identifies which geographic region the host is in, as well as the class of the network. There are five classes of IP addresses, and Internic assigns the IP addresses used on the Internet. The five classes and their corresponding numbers are:

Class A – A class A address has a first octet number between 1 and 126 and can support a network with 16,777,216 hosts. Very few of these addresses have every been assigned.
Class B – A class B address has a first octet number between 128 and 191 and can support 65,536 hosts. This is the most widely used class of addresses and it is now very hard to obtain one from Internic.
Class C – A class C address has a first octet number between 192 and 223 and can support 254 hosts per network.
Class D – A class D address begins at 224 and ends at 239; it is not widely used as these addresses are reserved for sending multicast messages. One day these addresses may be used for the Emergency Broadcast System.
Class E – These addresses begin with 240 and end at 255. These addresses are reserved for experimental use.

 

Binary

   Binary numbers is the basis that all computers communicate on. It is very important to understand how to transfer binaries into standard numbers for this exam. When you look at an IP address you see the format   xxx.xxx.xxx.xxx Every three digit section is called an octet.

Each octet of an IP address corresponds to a 8 bit binary number that the computer can understand. That is why the maximum octet number in an IP address is 255. Below is a table of the more common octet numbers and their corresponding binary number:

 

Octet Number

Binary Number

1

00000001

2

00000010

4

00000100

8

00001000

16

00010000

32

00100000

64

01000000

128

10000000

By adding the above numbers together you can obtain a result of any value between 1 and 255.

 

 

Transfer Control Protocol

TCP is the part of the protocol suite that provides the functionality for packets and error checking. Embedded in each packet are instructions the receiving computer uses to reassemble the data. Also, this protocol does the error checking if one packet is lost or corrupted then, this protocol issues a request to the host computer to send the data again.

 

Below is a list of commands and applications associated with the TCP/IP suite. Some of these are used primarily with Unix, but they are good to know anyway.

Ping – Typed from the command line, this protocol allows you to test connectivity between computers
SMTP (Simple Mail Transfer Protocol) – Allows the exchange of e-mail between two computers
FTP (File Transfer Protocol) Allows the transfer of files between two computers
Whois – Provides information about a host on a TCP/IP network
Telnet – Application that allows remote access and administration of a computer
Finger – Provides information about a particular user on a TCP/IP network. ( there are some Windows versions, but this is primarily used on UNIX machines)
TraceRoute- Used to determine the route a packet takes between two computers
Route Print – Used when troubleshooting to determine if TCP/IP is configured properly
Arp(Address Resolution Protocol)- Provides information on which computers your client computer has connected to recently( good for identifying the source of broadcast storms)
SNMP (Simple Network Management Protocol) Used when monitoring a network
Winipcfg – Used on Windows 95 / 98 computers to view TCP/IP configuration information
Ipconfig – Used on Windows NT computers to view TCP/IP configuration information
DHCP – Dynamic Host Configuration Protocol. As a network administrator you need to understand this. I found a great DHCP Article on Tech Republic, you may want to check out. You have to register to read it but it is well worth your time.

These are some of the more common commands used with this protocol suite