Re: Please Clarify Doubts regarding RGT & RNC

[Maarten Vissers <mvissers@lucent.com>]

All,

I would like to add a few notes to two items in the ongoing
correspondence:


>...
>         It should be noted that the LSP isn't necessarily used 
> to setup a network connection between two trail terminations. It
> can also be used to setup sub-network connections. In the later
> case you don't know the payload ID and specific GPID.

I am trying to generalise the layer network description from the control
plane's viewpoint (CP_LN). So far, I have come up with the following
udnerstanding:
  * CP_LN consists of a Fabric, a set of Ports and a set of Links
  * The set of Ports can be divided into 4 subsets:
    - towards a client LN (vertical relationship - top),
    - towards a server LN (vertical relationship - bottom), 
    - towards a user network (horizontal relationship - bottom) 
      providing native layer network service 
    - towards a user network (horizontal relationship - top)
      providing a circuit service.

	Example - native service is e.g. a VC-4 switched connection
	in a VC-4 layer network (VC-4 subnetwork connection is set-up);
	circuit service is e.g. a 140 Mbit/s (E4) switched circuit via
	a VC-4 layer network (VC-4 trail is set-up).


			to/from client CP_LN
			     ||||||||
     ------------	     --------
   --| UN_LN_CS |	     |      |
     --------|  |	     | C_LN |
             |  |	     |      |	
             ----	     --------
	       |		|
         -------------------------------------------
         |                  Fabric                 |
         -------------------------------------------
	     ||||	     ||||||||
             ----	     --------
   	     |  |	     |      |
     --------|  |	     | S_LN |
   --| UN_LN_NS |	     |      |
     ------------	     --------
				 |

		Figure 1 - 4 port types and fabric in CP_LN

The C_LN port provides a link to a client layer network.
The S_LN port provides a link to this layer network.
The UN_LN_CS port provides access to a circuit of a client layer
network.
The UN_LN_NS port provides access to a connection in this layer network.


			     ||||||||
  (A)-------------	     --------
   --| TTP       |	     |      |
     --------|   |	     | TTP  |
             |TTP|	     |      |	
             -----	     --------
	   (A')|	     (C)|                        |   |   |
         -------------------------------            -------------------
         |           Fabric            |            |     Fabric      |
         -------------------------------            -------------------
	 (B')||||	  (D)||||||||   <-- Link -->   ||||||||    |||
             ----	     --------    Connection    --------    ---
   	     |  |	     |      |    End Points    |      |    | |
  (B)--------|  |	     |      |                  |      |    | |
   --| USER LINK|	     |      |                  |      |
     ------------	     |      --------------------      |
			     |     (Topological) LINK         |
                             ----------------------------------

			Figure 2

Looking at figure 2, there are 4 interfaces to the Fabric: A (via A'), B
(via B'), C and D. Connections in a switched (sub)network can be made
between many combination of these 4 interface types in the ingress node
and egress node of the switched network:


* A - A ==> this includes a trail
* B - B ==> this is a subnetwork connection between UNIs (CTP - CTP)
* C - C ==> this is a trail (TTP - TTP)
* D - D ==> this is a subnetwork connection between NNIs (CTP - CTP); 
	    this is also an LSP as far as I understand

[hybrid connections:]
* A - B ==> trail termination (TTP) at one end and CTP at other end
* A - D ==> TTP - CTP
* B - A ==> CTP - TTP
* B - C ==> CTP - TTP
* B - D ==> CTP - CTP
* C - B ==> TTP - CTP
* C - D ==> TTP - CTP
* D - A ==> CTP - TTP
* D - B ==> CTP - CTP
* D - C ==> CTP - TTP

CTPs may have additional functionality to support:
- non-intrusive monitoring
- tandem connection monitoring; this may be a single instance or
  multiple instances of tandem connection monitoring. E.g. at a UNI
  there might be a Service Provider TCM and a Network Operator TCM
  endpoint active. At a NNI there might be a Network Operator TCM 
  endpoint active. Other cases may also exist.

> 
> >  I don't know if there is other trail termination info that
> > should be signalled between the endpoints.
> > Do you have any thoughts on this?
>         [Juergen Heiles]  Tandem Connection Monitoring and setup of
> the TCM level might need some thoughts. Do we want to setup TCM
> automatically via LSP signaling at all?

TCM endpoints may be required for SLA verification (holding time of
transport connections is rather long compared to voice connection), or
for detecting the trigger condition(s) (SF, SD) for restoration.

TCM endpoints for SLA verification may already be activated during
installation of the interface port unit and/or fiber connection. If not,
then the control plane must do it.

TCM endpoints for restoration may reuse the TCM endpoints for SLA
verification. If not, then the control plane must activate those.

Besides the possible activation of TCM endpoints, a trail or tandem
connection monitored connection set-up will have to perform the
following actions:

- read the MI_TxTI (transmitted trace id) value from the TT_So function
at the "A" end, send this TTI value to the "Z" end and assign it to the
MI_ExTI (expected trace id) signal in the TT_Sk function. MI_TIMdis is
then set to "false" and MI_TIMAISdis is also set to "false".
- read the MI_AcTI signal once the connection is set-up to confirm the
correct connection is created. If AcTI is ExTI then inform the other
end. If this end also has AcTI is ExTI the connection is setup
completely and now the reporting modes at each end can be set: either
set the MI_TPmode to the value "MON", or set the ARCmode for the TT_Sk
to the value "ALARMED".
- if the connection being setup is a trail, a check of the adaptation
function types at each end should also be included.
- if the connection is half of a trail/tandem connection then
  * if "A" end there is a TT_So, the TTI value must be handed-off at "Z"
    to the next switched network
  * if "Z" end is a TT_Sk, the TTI value must be incoming at the "A" end
    and if not available (i.e. "void"), then the MI_TIMdis signal is set
    to "true" (to disable trace id mismatch detection).

As such, the TTI read from MI_TxTI must be signalled, and detection of
MI_AcTI = MI_ExTI is to be signalled.

Some similar actions are necessary when there is a non-intrusive monitor
connected to the connection.

Regards,

Maarten
begin:vcard 
n:Vissers;Maarten
tel;cell:+31 62 061 3945
tel;fax:+31 35 687 5976
tel;home:+31 35 526 5463
tel;work:+31 35 687 4270
x-mozilla-html:FALSE
org:Lucent Technologies Nederland;NA&CPSE
version:2.1
email;internet:mvissers@lucent.com
adr;quoted-printable:;;Botterstraat 45=0D=0A=0D=0A;1271 XL Huizen;;;The Netherlands
fn:Maarten Vissers
end:vcard