Re: [Detnet-dp-dt] a quick look at the replication/discarding case for detnet flows

Balaazs,

I've been thinking about this, but still not sure I understand the issues.

On 2017-01-17 23:26, Balázs Varga A wrote:
> Hi Loa,
>
> Two comments/discussions points for the weekly call:
>
> 1, P nodes (recognizing detnet-flows)
> P nodes have to recognize detnet-flows in order to use the allocated resources (e.g., QoS) during the
> forwarding of packets belonging to a given detnet-flow. We have two options:
> (i) P nodes recognize "d-pw" label to identify the detnet-flow. This would mean that P nodes have to
> walk through the whole label stack.

No, I don't think the P nodes needs to look at anything other than
transport label. The required QoS is assigned to the transport label
at setup time, at forwarding only one label needs to looked at.

And I think also that "d-pw" label value must be globally unique

It needs to be unique within the replicas of the replicas and the
the incoming interface of the T-PW.

> in the network to make life simple! That can be a scalability issue ...

It is certainly a scalability issue, and it would have an effect all
traffic in the network and all levels.

> (ii) We are using different LSP per detnet-flow. This would make identification more simple for P nodes
> but raises scalability issues for the LSPs.
>
> 2, Distinguishing replica flows (required for OAM purposes)
> Let's assume that "node1 and node5" and "node2 and node6" are collapsed. I have created slide13
> based on your slide7 (collapsed nodes denoted as node15 and node26). As a result the upper and
> lower MS-PWs shares the first and second hop link when leaving nodeA.

yes, and the issue

  Node15 cannot distinguish
> the two replica flows only using the "d-pw" label.

Is the P-node really supposed do anything about the replicas than
forwarding them?
  We should consider to use different "d-pw" label
> value per "detnet-flow segment". Such a scenario is shown on slide14. Using different "d-pw" labels
> may furthermore obsolete the PW-labels ...

I think I lost you here, if we use díferent labels, how will the nodes 
know they are replicas? And P nodes only look at the transport label.

/Loa
>
> Cheers
> Bala'zs
>
> -----Original Message-----
> From: Detnet-dp-dt [mailto:detnet-dp-dt-bounces@ietf.org] On Behalf Of Loa Andersson
> Sent: Saturday, January 14, 2017 5:29 AM
> To: detnet-dp-dt@ietf.org
> Subject: [Detnet-dp-dt] a quick look at the replication/discarding case for detnet flows
>
> Folks,
>
> I took a quick look at the replication / discarding case we talked about at the meeting this week.
>
> Please remember that so far this is not a proposal, just a demonstration that there is at least one way of doing it.
>
> What is there can bee revised, improved, changed or discarded if we find something better. Take a look a send comments.
>
> /loa
>
> Warning: I have a grave case of dyslexia, so any odd or inventive spelling is pure accident.
>
>
>
> _______________________________________________
> Detnet-dp-dt mailing list
> Detnet-dp-dt@ietf.org
> https://www.ietf.org/mailman/listinfo/detnet-dp-dt
>

-- 

Loa Andersson                        email: loa@mail01.huawei.com
Senior MPLS Expert                          loa@pi.nu
Huawei Technologies (consultant)     phone: +46 739 81 21 64