|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Now that you have learned a little something about packet switched
networks, your next questions may be, How does the Internet know where the
data is going? How can a whole file be moved reliably from one place to
another without error, despite the fact that this file is broken down
into packets and transmitted over an unreliable network? The answer to
these questions is the TCP/IP protocol, which is the internet standard for
handling these problems.
An analogy that is used quite often to describe this process is the U.S.
Postal Service. The Postal Service is considered a packet switched
network, since there is no dedicated piece of network. If you want to send
something, it is mixed in with everyone else's stuff, put in a pipe,
transferred to another Post Office, and sorted out again. Although the technologies are completely different, The Postal Service is a surprisingly accurate analogy and will be used throughout this section of the tutorial.
Imagine that telephone lines and Ethernets are equivalent to the trucks and planes of the Postal Service. They are the means by which mail is moved from one place to another. The routers are postal substations; they make decisions about how the data (packets) shall be routed, just like the postal substation decides how to rou te the envelopes containing mail. Routers, like substations, don't have a direct link to all other routers. When you send a letter, your local post office forwards it to a substation; the substation forwards it to another substation; and so on, until it r eaches its destination. Each substation only needs to know which substations it has access to and what is the next closest substation to the destination. Similarly, the internet router looks at where your data is going and decides where to send it next.
INTERNET PROTOCOL (IP) Going back to the Post Office analogy, the Internet Protocol works like an envelope. Some addressing goes at the beginning of your message, which gives the network enough information to deliver the packet of data.
Internet addressing consists of four numbers each less than 256. The numbers are separated by periods and look something like this:
193.211.48.6
The address is made up of multiple parts, the first part tells the internet router what network the receiver is a part of. Remember, the internet is a network of networks. The second part tells the network which computer should receive the packet. Again, using the Postal Service analogy, consider the address of 50 Main Street, Fairfax, VA. The "Fairfax, VA" portion of the address is like the network address, it gets the package to the receivers local Post Office. The "50 Main Street" portion is like the host address, it identifies a particular mailbox within the local Post Office's network. Information that is transmitted across an IP network is broken down into pieces, called "packets". These packets are normally between 1 and 1500 characters long. One reason for this is that it prevents any one user from monopolizing the network for a lon g period of time.
The Internet Protocol in itself, contains all the information necessary to transmit a packet from one computer on a given network to a computer on a different network. However, this process would not be very friendly nor would it be efficient. Several major problems would arise that would need to be dealt with: 1) Most information transfers are longer than 1500 characters. It would be inconvenient if the Post Office would only carry postcards and nothing larger. 2) Packets may arrive out of order. If you were to mail several postcards in successive days, The Post Office does not guarantee that the postcards will travel the same route and arrive in the same order that they were sent. 3) Packets can get lost. The Post Office occasionally loses a letter and networks occasionally lose packets. Unlike the Post Office, networks can deal with deal with problems such as these. The next layer of the network, TCP, does just that.
TRANSMISSION CONTROL PROTOCOL (TCP) Say for example, you wish to send a book, but the Post Office will only handle letters. You could rip each page out of the book, place each page in a separate envelope, address it, and put them all in a mailbox. The recipient would have to make sure all t he pages arrived, put them back in the right order and paste them back together. This is what TCP does.
TCP takes the information you wish to send and breaks it into smaller chunks of data, namely; packets. It numbers each packet so that receipt can be verified and the data can be reassembled in the proper order. In order to pass this network across the net work, TCP places the data into an envelope of its own which has the number "written on it". Then this TCP envelope is placed in the IP envelope and sent across the network. On the receiving side, A TCP software package collects the packets, extracts the data, and puts it back into the original order. It checks to see if packets are missing and asks the sender to retransmit the missing packets if necessary. Once all the data has been collected and placed in the correct order, it passes the data onto the application program using its services. During this transmission packets can get lost and even changed. Fortunately, TCP can also handle this. When it places the data into an envelope it calculates something called a checksum. A checksum is a number that allows the receiving TCP to detect error s in the packets. When the packet arrives, the receiving TCP calculates what the checksum should be and compares it to the one received. If they don't match, an error has occurred and the receiving TCP discards the packet and requests a retransmission.
 |
 |
]
|
