Just spotted your post and thought I’d add a few ‘first person’ notes from the front lines which I’ve embedded inline below.
> Is TRILL a “terrible idea?” Does Shortest Path Bridging rule? I have no clue.
All protocols are compromises and standards based ones especially so. They both have aspects which I personally consider ‘terrible’ One of my first rules of protocol and network design however is never to love or hate a protocol. Just use what works at the cost points you require. Unfortunatley that does not make for very exciting blog posts.
> Vendors and pundits often gloss over the intricate differences between the two standards,
> probably because most of us in the media lack the technical knowledge to grasp the finer points involved.
The two most important differences are how packets are forwarded and how OA&M works. The IEEE solution fires packets like a rifle, targetting a very specific path while the IETF solution is more like a shot-gun. For OA&M the IEEE solution re-uses 802.1ag and Y.1731 .. for better or worse, while the IETF solution prefers something new , for better or worse. The IEEE forwarding solution will offer better control and runs on many current boards/asics while the IETF solution will offer better spread when hop counts get above 4 or so. Its a bit like comparing MPLS (802.1aq) to IP (TRILL) forwarding. Talking with customers we find some that don’t want control and want the network to be totally transparent, others demand control. For example the customer I spent all afternoon with today was delighted in the symmetry and positively hate trying to debug 1/2 failed connections. Therefore to address all customers will require both forwarding modes in one protocol, or both protocols. Intelligent vendors will listen to their diverse customer requirements and not fall in love or hate one or the other method. Some customers don’t mind spending money on new 40G cards if the cost brings them a new value while others naturally want to squeeze the maximum out of their investments, seems very reasonable to me.
> After reading through the slides, I really wish I was there for this talk.
It was recorded. There is a link at the bottom of the 802.1aq wikipedia link here:
which shows the Tutorial video recorded at Nanog and I’m pretty sure you can find the debate video linked in that same general NANOG directory.
> I wonder what the tone of this talk was like. Were the disagreements friendly and tongue-in-cheek,
> or are there really hard feelings on this issue?
Unfortunatley it was not as friendly as it should have been. I did review it and make a note to not rise to the bait as quickly next time but compared to some of the stuff I’ve seen around MPLS-TP (or the early MPLS v,s ATM debates) it was positively friendly
Regards Peter Ashwood-Smith (Huawei)]]>