As always, we started with the “what’s wrong with what we have right now, like using BGP as a better IGP” question, resulting in “BGP is becoming the trash can of the Internet”.
You can probably guess the next set of intro questions: "Do we need policies and traffic engineering in a high-speed data center fabric?"
Finally, we got to discussing OpenFabric, starting with how OpenFabric uses link-state database to modify IS-IS flooding behavior and thus reduces the total amount of flooding in a data center fabric. Next step: how do you know where it’s safe to reduce flooding? You have to figure out whether you’re a leaf or spine, and there are two ways in OpenFabric to solve that challenge.
Of course, we couldn’t possibly stay on track, so there was a brief diversion into virtual aggregation, application-level IP address huggers and local area IP address mobility with ARP… but quickly got back to more OpenFabric goodness including how it uses Segment Routing.
More to explore: