Re: [Srcomp] comments on REQ-8-26-HET-SID-LIST

Hi Darren, Ron,

I think that could be important for a brownfield deployment scenario but not as critical in a greenfield case.

Regards,

PSF

原始邮件

发件人：DarrenDukes(ddukes)
收件人：Ron Bonica;srcomp;
日 期 ：2020年09月30日 13:33
主 题 ：Re: [Srcomp] comments on REQ-8-26-HET-SID-LIST

-- 
Srcomp mailing list
Srcomp@ietf.org
https://www.ietf.org/mailman/listinfo/srcomp

Hi Ron

The metric states that a proposal permits the use of compressed and non-compressed segments within a segment list.

"A solution that supports compressed and non-compressed segments, and allows their combination within
 a segment list meets this requirement."

The benefits to deployment seem clear.  Some nodes support the compression solution while others do
 not yet, the service is incrementally upgradable.

Darren

--------------------------------------------------------------------------------

From: Srcomp <srcomp-bounces@ietf.org> on behalf of Ron Bonica <rbonica=40juniper.net@dmarc.ietf.org>
Sent: Tuesday, September 29, 2020 4:39 PM
To: srcomp <srcomp@ietf.org>
Subject: [Srcomp] comments on REQ-8-26-HET-SID-LIST 

Folk,

The rationale for this requirement are dubious.

You are proposing that SR ingress nodes must support the following SR encoding:

Legacy SRv6

Compressed

Hybrids containing Legacy encoding features and compressed encoding features

The rational are:

That this supports one particular transition strategies

That some nodes may not require compression

Other, more frequently deployed transition strategies are available. And what is the motivation for using classing encoding when compression is available?

                                                                       Ron

Juniper Business Use Only