Nice job on bit-serial.

The question of whether WANs are reliable and connection-oriented requires 
more explanation. I acknowledge Howard Berkowitz for teaching me this new 
way of looking at the question.

To understand PPP, ISDN, Frame Relay, X.25, and probably many other WAN 
protocols, it helps to recognize that these protocols have control and 
management planes that carry signaling and overhead information. The 
control and management planes are separate from the user plane that carries 
user data.

Think of the ordinary 7-layer model as the front (main plane). Now think of 
the model as being 3-dimensional, with a control plane and management plane 
in addition to the user plane in front.

The behavior of the control and management planes may be quite different 
than the behavior in the user plane. In fact, the control plane is probably 
connection-oriented and reliable, whereas the user plane is not.

The control plane handles call setup. Think of what happens when you make a 
telephone call. (Because WAN protocols have a telephone network legacy, it 
makes sense to use a telephone example.) When you lift the handset off the 
cradle of your telephone, the switch at the telephone company's local 
office senses that your telephone has gone off hook, provides dial tone, 
and accepts the numbers that you dial. This happens in the control plane. 
The interconnected switches that permit national and international calls 
also communicate with each other in the management plane, using complicated 
routing and administrative protocols such as Signaling System 7 (SS7). The 
user plane sends the actual telephone conversation.

This division of tasks also occurs in both LAN and WAN networks, although 
one difference is that the control plane protocols for WANs are often quite 
complex. (In LAN environments, people don't pay much attention to the 
control plane, although both ARP and IGMP could be considered control plane 
functions.) Control plane protocols in the WAN world include LCP in PPP, 
LAPD in ISDN, LAPF in Frame Relay, and LAPB in X.25.

NOTE
See the first chapter of Howard Berkowitz's "WAN Survival Guide" (John 
Wiley & Sons, 2001) for an elegant explanation of the different OSI planes 
and their functions.

Priscilla

At 04:20 PM 2/6/02, s vermill wrote:
>Steven A. Ridder wrote:
> >
> > is PPP connection-oriented with acks?   I thought it wasn't.
>
>Steven,
>
>No.  Yes.  I think the strict definition of "connection oriented" relates to
>layer 3 or above protocols (such as TCP).  Of course, PPP does negotiate a
>logical connection between two end points.  But it does so at layer 2.  And
>you are absolutely correct that PPP does NOT ack data packets with control
>characters in the way I described earlier.  I was merely refering to the
>"ACK" that you will see ONLY during the PPP negotiation phase (and only if
>you are using a sniffer or debugging ppp negotiation).  Bisynch protocols
>use a similar mechanism throughout data transmission that PPP only does
>during negotiation.  Sorry for adding to the confusion.  The terminology is
>bad enough without my help.
>
>Scott
________________________

Priscilla Oppenheimer
http://www.priscilla.com




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=34691&t=34629
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to