• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1068
  • Last Modified:

WAN and the OSI layers

Here's a question are teacher presented us the other day:  Why do Wide Area Networks ONLY focus on the physical and data link layers? I cant just outright ask the teacher for the answer because its for bonus points. So I figured i'd come to you guys. I understand WANs use the physical layer by sending electrical pulses over telephone lines and what not. And I understand WANs use the data link layer by using mac addresses of devices such as the CSU/DSU. But, why isn't the network layer included? Routers, a key part of WANs, work off of this layer to route information. Also, packet-switched networks within the telephone company are based on the PDU of the network layer, the PACKET! What am I missing here, or is this just a trick question that my teacher has used just to get the gears in my head turning? Please help!
0
Snap-0-Spine
Asked:
Snap-0-Spine
  • 5
  • 3
  • 3
  • +2
1 Solution
 
humeniukCommented:
Routers are layer 3 (network layer) devices.  The question may be in reference to something specific, but it's fairly vague - what does "focus on" mean?
0
 
tmcguinessCommented:
You know we aren't supposed to answer questions related to school work here, right? I'll just say something to give you something to think about.

Communications between networks is inter-network communications. Two networks that are separated by some distance(or not) but still communicate are having inter-network communications. Using the OSI model, there is only one layer where inter-network communications happen.

If somebody took the off-the-wall perspective that the network between the two connected networks does not take part in any inter-network communications, the person would not have considered the fact that not all inter-network communications  happens across a single hop.

People on 'ludes should not teach.

have fun!

0
 
Snap-0-SpineAuthor Commented:
So i should look at a WAN as the Point of Presence on two different networks and then everything inbetween those networks? Looking at the WAN that way, I might come close to seeing how the focus is on the physical and data link layers because the routers are more a part of the local networks. However, for data to get from one point of presence to another, the information has to be ROUTED throughout all the networks inbetween. Aaahhh, who needs bonus points anyway. Post your ideas just for the hell of it if ya'd like.
0
Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
Snap-0-SpineAuthor Commented:
And the one layer where inter-network communications happens is at the physical layer.
0
 
humeniukCommented:
And make sure to ask your teacher what he/she was thinking about.  I'd be interested to hear it.
0
 
wmilligaCommented:


The WAN focuses on the physical and data link layers, because it is the protocols running over these layers that are unique to the WAN environment. The layers above are NOT unique to a WAN. For example, you run TCP/IP over a WAN link and a LAN link. You would not run Frame Relay over a LAN link. It did blur with ATM in the past, but ATM is no longer used in a LAN environment, as it was with ATM-LANE or ATM-Classical IP and ATM 25Mbps. Hope this helps, and good luck with your studies!
0
 
humeniukCommented:
Good answer.
0
 
Snap-0-SpineAuthor Commented:
That definitely makes sense wmilliga. Yours will be the accepted answer, but if you don't mind I'd like to leave this question open for a day or two longer to see if anyone else chimes in. It'll also give me a chance to find out what exactly my teacher is trying to get at. Thanks for the great feedback!
0
 
Snap-0-SpineAuthor Commented:
I just remembered, wasn't X.25 also a network layer protocol used on WAN links?
0
 
wmilligaCommented:
Yes- but it was replaced by Frame Relay.

X.25 was well suited for the unreliable, analog lines of days past, due to the fact that it had alot of error correction mechanisms built in. As we went digital, X.25 went away, as the overhead associated with those mechanisms was no longer necessary.

Some other WAN technologies: SMDS, SONET/SDH, ISDN, MPLS, PPP, etc.
0
 
tmcguinessCommented:
Aha... I took a 'lude to try to help me think like him <just kidding - it was just a xanax ;^) > Seriously. I think I know what he is saying now. First I'd like to clear up the current discussion that physical and data link layers are unique to WANS. I know that my LAN runs on Ethernet which is in the data link layer.

OK... picture this:

I send some data (I'm just going to call it all data and not get confused over where it is packets and where it is frames)  from my workstation on LAN1 destined for Joe's workstation on LAN2. I use whatever program that utilizes the TCP/IP application layer stuff (maybe it's ftp, smtp, http it just doesn't matter for our purpose). The application layer hands it down to either ICMP, TCP, or UDP (remember TCP/IP really has the seven layers clumped into four). We'll ignore ICMP because it is special, but the data can be passed down to either tcp or udp which if memory serves me really map to layers 4-6 of the OSI model.  The transport protocols format the data add their headers, organize them and so on.  Once these protocols are through, they pass the data on to the layer three protocols which in our case we will just say it is IP. IP will stick the source and destination addresses, the source and destination port, and a good bit of other information into it's header it will also check to see if the address is part of the local network, if it is, it passes that information in the header along to the data link layer, if it is in a different network, it passes the address of the defult gateway in the header on to the data link layer.  The data link layer peeks at the information passed to it from IP and searches the arp table to see if it can find the device. It then does a broadcast to find the device. If the device is on a switch, the switch will pass the data along to the device. We'll talk about what happens when it reaches the host in a second which is what would happen if it was on the same network. But assuming that it was on a different network, the data would be passed on to the address of the default gateway (AKA router). The router peeks at the IP address to determine if the address is on any of the locally attached networks. If it is, it passes the data on to the appropriate layer two device on that network but before it passes it on, it takes the source MAC address out and replaces it with it's own addresss so the data will be able to find it's way back, and replaces the destination MAC address with the address of the layer two device which will receive the data. Assuming the router does not know where this device is, it will pass the data onto it's own default gateway, once again, replacing the source MAC with it's own and the destination MAC with the address of the desitnation router. If this router looks at the IP address and says "hey that is a network that is attached to me!" It will do a arp lookup to find the MAC address of the destination. It will replace the the destination address in the data's header with that address and replace the source MAC with it's own address. The data will then be passed on to the right machine.  The physical layer hands off the data to the data link layer that strips off it's header so it can keep track of whatever needs to be kept track of. The rest is passed up to IP where a whole bunch of stuff happens to establish a connection and it strips stuff off and then passes it up to let's say tcp which looks at the data puts it all back together and can then pass it up to the application needs to use it. Of course there is a whole bunch of jibber jabber going on that I am not going to go into.

Anyway... what your instructor is saying is that nothing happens with data at any protool layers above the Data Link. I have never thought of it in those terms but I can see where from this perspective where I could go along with that argument.
0
 
na_venkyCommented:

Data link layer : it is very simple,the layer name itself tells you that  this layer is ment for DATA and the link ,  how data is formated and present to the link(physcial link or physical layer),

LAN DL layer
Data formating  means that data can be for ethernet II ,novell network, or FDDI network. each network uses a diffrent framing methods ex FDDI ,SNAP , 803.2  etc .

WAN DL layer
when your transmitting the data on diffrent WANs for example FrameRelay ,ATM ,Asycronous (Dial UP, you use diffrent kind for data formatting or( correct term is) data encapsulation .
when u transmitt over framerelay network the dta will have frame relay header and trailer.
similarlyit applies to all networks.

Datalink layer is very much important  to both LAN and WAN

tell you instructor  about these things

 

0
 
tmcguinessCommented:
Boy did that last paragraph I wrote not make sense. Your instructor meant that nothing happens with the data at any protocol above the data link layer in WAN communications....
0
 
Snap-0-SpineAuthor Commented:
i'm sorry i didnt check this thread recently. everyone of you guys have been incredibly helpful and according to my teacher correct. don't worry, i didnt take credit for the work of others, i actually showed her the thread. the whole class, including the teacher, have been enlightened. i cannot thank you all enough!
0

Featured Post

VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

  • 5
  • 3
  • 3
  • +2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now