Hi vergita-ga,
Let's first consider a data packet:
A data packet is generally designed to acommodate network protocols
(standards defining how source and destination should communicate).
Most networking protocol, such as TCP/IP, are designed as "layered"
protocols where each layer performs a specific function. The best way
to understand this is by considering what happens to data as it passes
through the layers between source and destination.
Consider an application on System A that is trying to send data to
another application on System B. It is not reasonable for these two
applications to communicate directly; they must follow a protocol. We
start with the actual data being sent from A. The application layer
appends a "header" to the data and sends the information down to the
next layer in the protocol. Similarly, all layers but the last one
append their own header. By the time the packet reaches the bottom
layer of system A, it has several headers appended (in order of layer)
to the actual data - this entire packet is called a PDU (Protocol Data
Unit). The purpose of all these headers vary by protocol but generally
include all sorts of control information so that the packet is
interpreted properly by the corresponding layer at System B. After the
PDU has been transferred to System B, each layer (from the bottom up)
proceeds to strip off one header, interpret it, and pass the remaining
portion of the PDU to the next higher level. At the top level, the
application receives a PDU with the final header and uses it to
understand the data.
The following is a link that describes the X.25 protocol, and although
you may not want to get tied into the details of this page, consider
the two diagrams that outline the layered networking approach and the
structure of a data packet:
http://www.rad.com/networks/1996/x25/x25.htm
Now for the next part of your question, circuit switching vs. packet
switching:
CIRCUIT SWITCHING:
"A type of communications in which a dedicated channel (or circuit) is
established for the duration of a transmission. The most ubiquitous
circuit-switching network is the telephone system, which links
together wire segments to create a single unbroken line for each
telephone call.
http://www.webopedia.com/TERM/C/circuit_switching.html
PACKET SWITCHING:
"Refers to protocols in which messages are divided into packets before
they are sent. Each packet is then transmitted individually and can
even follow different routes to its destination. Once all the packets
forming a message arrive at the destination, they are recompiled into
the original message."
http://www.webopedia.com/TERM/P/packet_switching.html
The last two links also provide some of the common scenarios when each
method is most preferred:
"Circuit-switching systems are ideal for communications that require
data to be transmitted in real-time. Packet-switching networks are
more efficient if some amount of delay is acceptable."
"Circuit-switching networks are sometimes called connection-oriented
networks. Note, however, that although packet switching is essentially
connectionless, a packet switching network can be made
connection-oriented by using a higher-level protocol. TCP, for
example, makes IP networks connection-oriented."
So packet switching should be considered over circuit switching when
the following is true:
1. Data is transferred in short bursts rather than continuously
2. Network topology is not constant - packet switching can use dynamic
routing to eliminate possible transfer failures.
3. It is not useful to set up and tear down a connection for each
transfer
4. There are multiple potential paths through the network that can be
utilized simultaneously
5. More bandwidth is required than a network under circuit switching
can handle - packet switching uses queuing at the intermediate nodes
to alleviate this problem.
I hope you have found this useful...if you have any problem
understanding the information above please post a clarification and I
will respond promptly :)
Cheers!
answerguru-ga |
