Skip to main content

Shepherd writeup
draft-ietf-cdni-footprint-capabilities-semantics

1. Summary

The document shepherd is Francois Le Faucheur.  The responsible AD is Barry
Leiba. The requested RFC status is Informational.

This document describes the semantic of a footprint and capability
advertisement function usable by interconnected CDNs allowing downstream CDNs
(dCDNs) to advertise capabilities and the coverage footprint of those
capabilities to upstream CDNs (uCDNs). This function is to be supported by the
CDNI Footprint & Capabilities interface (FCI) identified in the CDNI Framework
(RFC 7336). The FCI is intended to help an uCDN decides whether or not to
delegate requests to a given dCDN.

The WG had extensive discussions about the potential scope of the CDNI
Footprint & Capabilities interface (FCI); this document provides a useful
record of the key aspects of that discussion and the conclusions as to the
scope agreed by the WG for FCI. Based on this, the document also captures the
semantics of the FCI, defines a Base Advertisement Object, defines the
necessary registries and provides guidelines on how these registries may be
extended in the future. The document does not define, nor select, a specific
protocol to realise the FCI interface.

2. Review and Consensus

The document contains historical information discussing and defining the scope
of FCI (e.g., non-real-time updates, with no attempts to prevent CDNs from
lying about footprints, and without requiring CDNs to divulge topology or
capacity information).  There was a long history of debate over these issues,
but consensus was reached.  This document defines the broad WG consensus on a
minimum set of capabilities to be advertised, and the WG consensus on a minimum
set of footprint types to support, with registries created for the future
extensibility of both.  The document also defines an abstract object for
defining new capabilities and footprints. This document is intended to provide
guidance for the FCI interface standards track documents.

An independent review was performed by Iuniana Oprescu; all comments were
addressed.

A large discussion took place over logistical issues around how to handle
common references between FCI and CDNI Metadata, CDNI Logging, and CDNI
Redirection interfaces. A broad consensus was reached after many small design
team meetings were held to consider options.  The current IANA registry and
media type structure is the result.

As document shepherd, I performed a final review including IANA considerations
and normative references, as well as citing editorial and consistency issues;
all comments were addressed.

3. Intellectual Property

Each author has confirmed conformance with BCP 78/79.

There was one IPR disclosure from Juniper Networks.  The disclosure was
announced on the list with only one response on the list stating that the
RAND-like terms seemed reasonable. The disclosure was also discussed in the WG
meeting at IETF94.  The chairs, as individuals, agreed that the RAND-like terms
seemed reasonable and that they did not see any reason to hold up the document
and, as chairs, proposed to go ahead with current plans to progress the
document, and asked if anyone had any objection. There were no objections from
the WG.

4. Other Points

There are no downward references in the document.

The IANA Considerations registers three new CDNI Paylod Types. One of the
Authors is lined up as the designated expert reviewer for the CDNI Protocol
Types registry.

The IANA Considerations also creates a new registry for redirection modes,
providing sufficient guidelines to IANA. One of the Authors is lined up as the
designated expert reviewer for the CDNI Capabilities Redirection Modes registry.
Back