Hi.
I was reading thru BD 10k press releases(I know, more prone to fantasy
than technical references), and they suggested that BD 10k breaks with
the long Extreme history on forward-cache population and starts doing
pure topology-based L3 forwarding.
1) Is that true for plain routing, no ACLs, no PBR ?
2) Are there DoS controls in order to prevent routing entries with
incomplete adjacencies (i.e., CPU still need to ARP that address) to
overload the slow-path ?
3) What happens with ACLs, can it still do ASIC-only ? (With or
withour CLEAR-flow)
4) With CLEAR-flow on, will it punt those packets to the CPU ?
Tks,
Rubens
_______________________________________________
extreme-nsp mailing list
extreme-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/extreme-nsp
I was reading thru BD 10k press releases(I know, more prone to fantasy
than technical references), and they suggested that BD 10k breaks with
the long Extreme history on forward-cache population and starts doing
pure topology-based L3 forwarding.
1) Is that true for plain routing, no ACLs, no PBR ?
2) Are there DoS controls in order to prevent routing entries with
incomplete adjacencies (i.e., CPU still need to ARP that address) to
overload the slow-path ?
3) What happens with ACLs, can it still do ASIC-only ? (With or
withour CLEAR-flow)
4) With CLEAR-flow on, will it punt those packets to the CPU ?
Tks,
Rubens
_______________________________________________
extreme-nsp mailing list
extreme-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/extreme-nsp