Re: [mpls] Poll 2: New SPL vs re-purposed ELI for carrying Network Action Indicators

Hi Zafar

There are feature limitations with Merchant silicon using Broadcom chipset
which I have experienced in the past which is common across all vendors
using Broadcom chipset.

The tradeoff with Broadcom is you get high port density at a tradeoff of
lack of feature richness and I believe QOS is generally lacking with
Broadcom in my experience.

We need to test and have it work on vendor proprietary NPU such as ASR 9900
platform or NCS 6k NPU based platform.

For interoperability I would think we want it to work on Merchant silicon
but also on vendor proprietary NPU platforms.

In my experience Broadcom chipset is good positioned for P nodes packet
pushers but the heavy  lifting done by PEs should be on vendor proprietary
NPUs for feature richness.

Kind Regards

Gyan

On Thu, Jun 16, 2022 at 10:33 PM Zafar Ali (zali) <zali=
40cisco.com@dmarc.ietf.org> wrote:

> Tony, Kireeti,
>
>
>
> So “I do not want to modify my current implementation” is the argument to
> stop people interested in Bruno’s draft from using it?
>
> If such HW can/ need be modified to implement the new complex encoding
> scheme you proposed, why it cannot implement simple encoding in Bruno’s
> draft?
>
>
>
> Please also look at
> https://datatracker.ietf.org/doc/html/draft-decraene-mpls-slid-encoded-entropy-label-id-04#section-4.4
> and
> https://datatracker.ietf.org/doc/html/draft-decraene-mpls-slid-encoded-entropy-label-id-04#section-4.5
>
>
>
> Thanks
>
>
>
> Regards … Zafar
>
>
>
>
>
> *From: *Tony Li <tony1athome@gmail.com> on behalf of Tony Li <
> tony.li@tony.li>
> *Date: *Thursday, June 16, 2022 at 3:10 PM
> *To: *"Zafar Ali (zali)" <zali@cisco.com>
> *Cc: *Kireeti Kompella <kireeti.ietf@gmail.com>, "
> bruno.decraene@orange.com" <bruno.decraene@orange.com>, "mpls@ietf.org" <
> mpls@ietf.org>, "pals-chairs@ietf.org" <pals-chairs@ietf.org>, "
> mpls-chairs@ietf.org" <mpls-chairs@ietf.org>, DetNet Chairs <
> detnet-chairs@ietf.org>
> *Subject: *Re: [mpls] Poll 2: New SPL vs re-purposed ELI for carrying
> Network Action Indicators
>
>
>
>
>
>
>
>
>
> The draft has been implemented for the exact (SLID) use-case on hardware
> platforms based on Broadcom Jericho2 family of ASIC
> Please take a look at
> https://datatracker.ietf.org/doc/html/draft-decraene-mpls-slid-encoded-entropy-label-id-04#section-7
>
> No such issue was noted
>
>
>
>
>
> And Jericho2 is the only thing that counts in the entire world?
>
>
>
> T
>
>
> _______________________________________________
> mpls mailing list
> mpls@ietf.org
> https://www.ietf.org/mailman/listinfo/mpls
>
-- 

<http://www.verizon.com/>

*Gyan Mishra*

*Network Solutions A**rchitect *

*Email gyan.s.mishra@verizon.com <gyan.s.mishra@verizon.com>*

*M 301 502-1347*