[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: Need for hierarchical SONET/SDH LSPs.
Venkat,
see my comments below.
Juergen
+AD4- -----Original Message-----
+AD4- From: Venkat Dabbara +AFs-mailto:venkat.dabb+AEA-wipro.com+AF0-
+AD4- Sent: Thursday, April 18, 2002 7:36 AM
+AD4- To: Maarten Vissers+ADs- Vinay Vernekar
+AD4- Cc: ccamp+AEA-ops.ietf.org+ADs- gmpls-ops+AEA-mplsrc.com
+AD4- Subject: Re: Need for hierarchical SONET/SDH LSPs.
+AD4-
+AD4-
+AD4- hai,
+AD4-
+AD4- I have a question regarding the Sonet/SDH traffic bandwidth encoding
+AD4-
+AD4- This is in continuation to the scenario which vinay has
+AD4- mentioned in his mail.
+AD4- Consider three LSRs A, B and C as shown below.
+AD4-
+AD4- A---------------B---------------C
+AD4-
+AD4- Now at node A i initiate an LSP creation with SONET/SDH
+AD4- traffic parameters with elementary
+AD4- signal type as VC-4 and all other fields as zero and MT field
+AD4- as 1 ofcourse. Precisely i am trying to
+AD4- create a VC-4 lsp.
+AD4-
+AD4- Some doubts on the above scene
+AD4-
+AD4- 1+AD4- Now assume that B supports some sort of concatenation.
+AD4- Will it matter in any way to the
+AD4- LSP creation initiated at node A when the request reaches
+AD4- node B and further ??
+AD4-
No, as you setup a VC-4 LSP concateantion doesn't matter.
+AD4- 2+AD4- Now when the label request reaches the node C, the
+AD4- egress, and if that node satisfies the
+AD4- bandwidth asked for, then what label will it give to
+AD4- node B with respect to SUKLM schema
+AD4- if node C is capable of switching STS-1 level signal.
+AD4- Will it be with S+AD0-1, U+AD0-1 and others 0 ??
As you ask for a VC-4 or STS-3c-SPE only S is of interest. S can be any number between 1 and N for a STM-N line signal between B and C. This U would be 1 accordign to the 03 draft in order to indicate a VC-4 and KLM would be 0.
+AD4- If yes, it means that the label structure is guided by
+AD4- the type of signal being requested in the
+AD4- Sonet/Sdh bandwidth encoding.
Yes
+AD4-
+AD4- 3+AD4- Lastly when a label structure is formed, what sort of
+AD4- communication will GMPLS do with the link
+AD4- layer for creation of crossconnect ??
The communication with the cross-connent matrix is internal to the network element. It is not defined as it depends on your specific implementation. Basically the cross-connect matrix at B has to be set in such a way that the VC-4 time slot used for the LSP on the A-B link is connected to the VC-4 time slot used for the LSP on the B-C link.
+AD4-
+AD4- Please do clarify there doubts
+AD4-
+AD4- thanks in advance
+AD4-
+AD4- regards
+AD4- venkat
+AD4-
+AD4-
+AD4-
+AD4-
+AD4-
+AD4-
+AD4- -----Original Message-----
+AD4- From: Maarten Vissers +ADw-mvissers+AEA-lucent.com+AD4-
+AD4- To: Vinay Vernekar +ADw-vinay.vernekar+AEA-wipro.com+AD4-
+AD4- Cc: ccamp+AEA-ops.ietf.org +ADw-ccamp+AEA-ops.ietf.org+AD4AOw-
+AD4- gmpls-ops+AEA-mplsrc.com +ADw-gmpls-ops+AEA-mplsrc.com+AD4-
+AD4- Date: Wednesday, April 17, 2002 1:40 AM
+AD4- Subject: Re: Need for hierarchical SONET/SDH LSPs.
+AD4-
+AD4-
+AD4- +AD4-Vinay,
+AD4- +AD4-
+AD4- +AD4-I assume that the signal on the fiber between A and B is an
+AD4- OC-12/STM-4 optical
+AD4- +AD4-signal (carrying the STM-4/STS-12 signal) and on the fiber
+AD4- between B and C it is
+AD4- +AD4-a OC-48/STM-16 optical signal (carrying the STM-16/STS-48
+AD4- signal). Nodes A, B
+AD4- +AD4-and C will have HOVC/STS switch fabrics which may support a subset of
+AD4- +AD4-VC-3/STS-1, VC-4/STS-3c, VC-4-4c/STS-12c, VC-4-16c/STS-48c
+AD4- swtiching+ADs- e.g. A
+AD4- +AD4-node may support VC-4 switching, where as B and C nodes may
+AD4- support VC-4 and
+AD4- +AD4-VC-4-4c switching.
+AD4- +AD4-
+AD4- +AD4-If all nodes support VC-4/STS-3c switching (and they
+AD4- typically do) and the
+AD4- +AD4-STM-N/STS-N links between the nodes also support VC-4/STS-3c
+AD4- link connections
+AD4- +AD4-(also typically the case), then you can setup a VC-4/STS-3c
+AD4- connection between
+AD4- +AD4-nodes A and C. The VC-4/STS-3c payload is totally
+AD4- unimportant for this
+AD4- +AD4-connection+ADs- any payload can be present.
+AD4- +AD4-
+AD4- +AD4-Regards,
+AD4- +AD4-
+AD4- +AD4-Maarten
+AD4- +AD4-
+AD4- +AD4APg- Vinay Vernekar wrote:
+AD4- +AD4APg-
+AD4- +AD4APg- Hi,
+AD4- +AD4APg- Can I have a SONET/SDH LSP that passes through a SONET ADM
+AD4- as its transit
+AD4- +AD4APg- LSR??
+AD4- +AD4APg- Consider a situation as shown below -
+AD4- +AD4APg-
+AD4- +AD4APg- A---------------B---------------C
+AD4- +AD4APg-
+AD4- +AD4APg- The link A-B is a STS-12 link, B does a byte-interleaved
+AD4- multiplexing of 4
+AD4- +AD4APg- STS-12 and forms a STS-48, thus B-C is STS-48.
+AD4- +AD4APg- Can a LSP for unstructured VC-4 be setup with A as
+AD4- ingress, B as transit and C
+AD4- +AD4APg- as the egress??
+AD4- +AD4APg- As per the draft +ACI-Framework for GMPLS based control of
+AD4- SONET/SDH networks+ACI- the
+AD4- +AD4APg- frames are not labeled instead it is the flow that is
+AD4- being labelled. It also
+AD4- +AD4APg- mentions that tunnels should be used when switching from a
+AD4- different signal
+AD4- +AD4APg- level i.e VC-12 to VC-4.
+AD4- +AD4APg- With this in mind, I would assume that a LSP as mentioned
+AD4- should be possible,
+AD4- +AD4APg- because the signal type remains same i.e VC-4 even the
+AD4- frames get interleaved
+AD4- +AD4APg- from STS-12 to STS-48.
+AD4- +AD4APg- Please correct me if I am wrong.
+AD4- +AD4APg-
+AD4- +AD4APg- Thanks in advance.
+AD4- +AD4APg- Regards
+AD4- +AD4APg- Vinay
+AD4- +AD4APg-
+AD4- +AD4APg-
+AD4- +AD4APg- Name: Wipro+AF8-Disclaimer.txt
+AD4- +AD4APg- Wipro+AF8-Disclaimer.txt Type: Plain Text (text/plain)
+AD4- +AD4APg- Encoding: 7bit
+AD4-
+AD4-