Re: [pim] draft-ietf-pim-multicast-lessons-learned-00.txt

["Stevens, Jim A Collins" <[e-mail]>]

Comments on draft-ietf-pim-multicast-lessons-learned-00.txt
The IETF datatracker status page for this Internet-Draft is:
https://urldefense.com/v3/__https://datatracker.ietf.org/doc/draft-ietf-pim-multicast-lessons-learned/__;!!MvWE!DcJM0ErjeQk9jqnrgTrMq0pvAgn6Kd15yP7sRi68ZyF-hEKN65lZtXMF63ecaNK_cLpw8wH9d1Mk0k68wCvwblHxS6oyRhGQ$ 

1.  In section 3.2, I agree with statement that "The main advantage of PIM-BIDIR is scaling. It scales well when there are many sources for each group such as with videoconferencing and many to many financial applications."  In fact, I use PIM-BIDIR in MANET networks with hundreds of sources per multicast group and the transmitting and receiving nodes join and leave very dynamically including moving between networks.  The fact that the shared tree do not rapidly update with the nodes joining and leaving is an advantage since the shared tree is not constantly changing.  This same advantage was noted in RFC 7287 "Mobile Multicast Sender Support in Proxy Mobile IPv6 (PMIPv6) Domains"

2.  Section 3.2 also states that " However, with the lack of source-based trees, the traffic is forced to remain on the inefficient shared tree."  For the MANET networks and MANET network of networks that I see, the PIM-BIDIR shared trees are efficient as long as good RP locations are chosen.  And in these MANET networks, the stability of the shared tree is important as nodes move around dynamically including between different MANET networks.

3.  Need to define TCAM which is used without being defined.  (by the way, I liked how you defined the various acronyms in section 2 up front.) 
  You should also put other acronyms like SD, SDR, AMT, etc in section 2 as well

4.  I recommend that RFC 8815 " Deprecating Any-Source Multicast (ASM) for Interdomain Multicast" and RFC 5110 " Overview of the Internet Multicast Routing Architecture "be referenced

Jim Stevens