Skip to main content

Shepherd writeup

draft-trossen-sfc-name-based-sff has been presented to the ISE for publication
as an Informaitonal RFC.

Service Function Chaining (SFC) utilises Service Function Paths (SFPs) to
navigate packets through a network and ensure that Service Functions (SFs) are
executed on the packets before their delivery to their addressed destination.
SFs may be instantiated at multiple locations in a network and a choice must be
made to direct the packets to the best instances. This document describes a way
to separate a logical SFP from the specific execution end points (i.e., SFs)
through a name-based system. The document sets out the necessary extensions,
additional functions, and protocol details to enable a Service Function
Forwarder to handle name-based relationships.

The Abstract will contain the caveat text

   This document presents InterDigital's approach to name-based service
   function chaining.  It does not represent IETF consensus and is
   presented here so that the SFC community may benefit from considering
   this mechanism and the possiblity of its use in the edge data

Implementations of some aspects of the solution have been developed and are
currently being trialed as part of the H2020-funded FLAME project in Bristol
(UK) and Barcelona (Spain). Early versions of those solutions were trialed as
early as 2017 in Bristol as part of the H2020-funded POINT project.

The authors brought this document to the SFC working group and presented it at
a couple of meetings, but the chairs decided that it sat outside the working
group charter and recommended taking it to the Independent Stream. One of the
chairs (Joel Halpern) also did a review of the document.

Reviews were also provided for the ISE by Dirk von Hugo and Andy Malis. And the
ISE also did a review. The document has been updated several times as a result.

The ISE also consulted with the ICNRG chairs to check for overlap. They
responded that they see this work as not part of the scope of the ICNRG and
have no objection to it being published as an Independent Submission.

There are two IPR disclosures against this document at and Counsel for the IPR holders are
considering an update to the license in order to specifically handle the case
where the document ends up as an RFC outside the IETF Stream.