Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

MSP vs SNCP

7,683 views
Skip to first unread message

Ank

unread,
Jul 3, 2007, 5:24:20 PM7/3/07
to
Dear All,

Could anyone please tell me the basic and pratical difference between
MSP and SNCP?
Also can someone tell me when can we use MSP and when can we go for
SNCP?


I know the practical set-up for both but couldn't figure out when to
use which one of these two. With advantages and disadvantages of
each.


Hope the reader would be clear with my query and will be able to
provide me the details.


Thanks for your inputs in advance.


Regards,
Ankur

Huub van Helvoort

unread,
Jul 3, 2007, 5:45:45 PM7/3/07
to
Hello Ankur,

You asked:

> Could anyone please tell me the basic and pratical difference between
> MSP and SNCP?

MSP = Multiplex section protection, because Multiplex section,
Regenerator section and Optical section are in general 1:1:1
MSP is in general protecting the fibers between two nodes.

SNCP = sub-network connection protection, this is in general
protection of (a section of) a path in a network and applied
to VC-n.

> Also can someone tell me when can we use MSP and when can we go for
> SNCP?

Physcal section and path, as described above.

> I know the practical set-up for both but couldn't figure out when to
> use which one of these two. With advantages and disadvantages of
> each.

MSP can only be applied between two adjacent nodes.
SNCP can be aplied on a path through a network, passing
many nodes.

Cheers, Huub.

--
reply to hhelvooort with 2 'o's
================================================================
http://www.van-helvoort.eu/
================================================================
Always remember that you are unique...just like everyone else...

Azhar

unread,
Jul 4, 2007, 1:53:25 AM7/4/07
to

SNCP is faster than MSP (less switching time).
MSP has 16 nodes limit where as no limit for SNCP.
Minimum protection unit of SNCP is T1 (or E1) where as for MSP is
STM-1 (not sure about SONET).
SNCP is always 1+1 (to my knowledge but Huub can verify it) where as
MSP can be 1+1 or 1:N.
MSP is implemented by switching matrix card where as SNCP is
implemented by tributary card.

Huub van Helvoort

unread,
Jul 4, 2007, 5:02:09 AM7/4/07
to
Hello Azhar,

You added:

>> Could anyone please tell me the basic and pratical difference between
>> MSP and SNCP?
>> Also can someone tell me when can we use MSP and when can we go for
>> SNCP?
>>
>> I know the practical set-up for both but couldn't figure out when to
>> use which one of these two. With advantages and disadvantages of
>> each.
>>
>> Hope the reader would be clear with my query and will be able to
>> provide me the details.
>>
>> Thanks for your inputs in advance.
>>
>> Regards,
>> Ankur
>
> SNCP is faster than MSP (less switching time).

The requirement is faster than 50 ms for each, it is up to
the implementation how fast it actually is.
MSP is normally faster because it covers only one section,
SNCP covers more sections and experiences more propagation delay.

> MSP has 16 nodes limit where as no limit for SNCP.

MSP has 2 nodes (yes the APS protocol has space for 16 but that is
irrelevant for MSP), MS-SPRing (the ring protection variant of
section protection) supports up to 16 nodes.

> Minimum protection unit of SNCP is T1 (or E1)

This should be VC-11 or VC-12 for SNCP

> where as for MSP is STM-1 (not sure about SONET).

STM-0 is covered as well, as are the SONET OC-x rates.

> SNCP is always 1+1 (to my knowledge but Huub can verify it)

SNCP can be 1+1, 1:1, n:m, (1:1)^n, see ITU-T rec. G.808.1

> where as MSP can be 1+1 or 1:N.

OK.

> MSP is implemented by switching matrix card

MSP can be implemented on the line card and in the HO
connection function (VC4 switching matrix).

> where as SNCP is implemented by tributary card.

SNCP is a function supported by the HO connection function
(VC-4) and LO connection function (VC-3, VC12, VC11)
see the functional models in ITU-T rec. G,783

ajaygoe...@gmail.com

unread,
Jul 16, 2012, 4:02:53 AM7/16/12
to
MSP is harware protection with a standby card while SNCP is a path protection.

On Wednesday, 4 July 2007 02:54:20 UTC+5:30, Ank wrote:
> Dear All,
>
> Could anyone please tell me the basic and pratical difference between
> MSP and SNCP?
> Also can someone tell me when can we use MSP and when can we go for
> SNCP?
>
>
> I know the practical set-up for both but couldn't figure out when to

anandku...@gmail.com

unread,
Mar 21, 2018, 11:16:58 PM3/21/18
to
Can anyone tell that we can implement MSP and SNCP in PTP , Ring or mesh typologies ...???
Is MSP applied between two nodes only and SNCP in entire ring to protect the services end to end ...???

Huub van Helvoort

unread,
Mar 22, 2018, 2:06:17 PM3/22/18
to
Hello Anand,
Did you read my response to Ankurs questions?
I explained the difference at that time (July 3 2007)

Also read ITU-T recommendation G.841 which describes these types
of protection.
In short:
-- MSP is section protection between two adjacent nodes
-- SNCP is path protection between two end-points (AKA linear protecton)
-- MS-SPRing is ring protection

Regards, Huub.

Ahmed Mustafa

unread,
May 15, 2018, 5:09:49 AM5/15/18
to
Hi Mr.Huub ,

about the switching timing in SNCP,MSSP,MSP what are the factors to determine the time?
and SNCP is dual fed and selective receiving why the passing through nodes effecting the time switching?

and another Q for MSSP , why is the limitation only for 16 nodes? is this only becuase k1,k2 4 bits, is there other reason ?

Huub van Helvoort

unread,
May 17, 2018, 11:16:39 AM5/17/18
to
Hello Ahmed,

You wrote:

> about the switching timing in SNCP,MSSP,MSP what are the factors
> to determine the time?

The switch time is measured from the time a defect (like signal fail)
has been detected until the service has been switched and is restored.

This includes transmission latency (5 ms per 1000 km), the node latency
20 microseconds per node, and the processing time (implementation
dependent.

> and SNCP is dual fed and selective receiving why the passing through
> nodes effecting the time switching?

The APS protocol processing time at both endpoints determines the
switching time.

> and another Q for MSSP , why is the limitation only for 16 nodes?
> is this only becuase k1,k2 4 bits, is there other reason ?

It is indeed because there are only 4 bits available.
It is not very likely that there are more than 16 nodes in a ring.

Best regards, Huub.

kunj.bih...@gmail.com

unread,
Jun 16, 2019, 3:53:03 PM6/16/19
to
Why only 16 node can be in msp??

Huub van Helvoort

unread,
Jun 17, 2019, 5:00:52 AM6/17/19
to
Hello kunj.bihari

You wrote:

> Why only 16 node can be in msp??

If you look at the APS bytes/bits you will notice that there are
only four bits allocated for the node ID. This limits the number
of node to sixteen.

More than 16 nodes on a ring will affect the response/switching
time which should be less than 50 ms.

even....@gmail.com

unread,
Jun 21, 2020, 7:04:18 AM6/21/20
to
----
@HUUB VAN HELVOORT
PLEASE BRIEF..
It is indeed because there are only 4 bits available.
> It is not very likely that there are more than 16 nodes in a ring.
BEST REGARDS

Huub van Helvoort

unread,
Jun 22, 2020, 10:34:28 AM6/22/20
to
Hello Even,

You wrote:

> @HUUB VAN HELVOORT
> PLEASE BRIEF..
>> It is indeed because there are only 4 bits available.
>> It is not very likely that there are more than 16 nodes in a ring.

Your question is too brief for me.

What is the information that you would like to see?
0 new messages