[phirephly]

When you want line rate forwarding across several Tbps of front panel ports, you need the packet pipeline to be able to make all the routing decisions without involvement from the OS. 8Bpps just doesn't give you time to be able to walk any kind of data structure in memory.

Running full internet tables on a x86 server where you can only get a few Gbps up to maybe a few dozen Gbps is much easier.

[cookiengineer]

Isn't this what BPF and XDP try to target with their SmartNIC offloading, like the ones supported from Metronome?

[phirephly]

Sure. And switch ASICs hang off a CPU as a PCIe device, so the line between smartNICs and low end ASICs start to blur. The key is hardware acceleration of some sort.