On Fri, Sep 29, 2023 at 12:54 PM Collider <large.hadron.collider@gmx.com> wrote:
This thread is utter amateur hour. I too would rather /32s be valid in the DFZ - but they're not, for good reason (worstcase scenario = circa 4 bln. routing table entries - no BGP hwaccel can swing that!).
Howdy, Actually, BGP can swing that. Routing involves two distinct components: the routing information base (RIB) and the forwarding information base (FIB). BGP is part of the RIB portion of that process. It's always implemented in software (no hardware acceleration). It's not consulted per-packet, so as long as the update rate is slow enough for the CPU to keep up and there's enough DRAM (which is cheap and plentiful these days) to hold the entire thing, there's no particular upper bound to the number of routes. The router then processes all of the RIBs for all of the routing protocols used on the router to find the best routes for any given destination and match them to the next hops to which packets should be sent. The results are stored in the FIB. This FIB is what's consulted for each packet passing through the router. The limiting factor is the FIB. The FIB is what is implemented with hardware acceleration on high-end routers in order to achieve large packet-per-second (PPS) numbers. It relies on storage hardware which is both faster and more expensive than DRAM. Consequently it has much less capacity to store information than DRAM. Currently shipping equipment intended for BGP backbone use can manage 1M to 2M routes in the hardware-accelerated FIB regardless of the amount of DRAM on the machine. Regards, Bill Herrin -- William Herrin bill@herrin.us https://bill.herrin.us/