tarom...@gmail.com [tarom...@gmail.com] wrote: > This is a really interesting thread. > > From my novice perspective, I wonder if the interrupt load actually makes a > difference on the performance of OpenBGPd on different hardware as bge or em. >
A higher interrupt load makes the CPU busy running the interrupt handler more. This doesn't matter as much with high end CPUs, but as the packet-per-second load increases, it helps. > I always assumed different NICs and drivers had different "behaviours", > characterized, for instance, by the features supported or interrupts per > second handled. > Sure, of course. Drivers are like any other piece of software, some are better, some have more features, some have problems, whatever. > My question, if not too offtopic, is if we can use the interrupt load as a > measure of NIC throughput. My uninformed guess is that its not. > No the interrupt load won't reflect real traffic while some type of interrupt mitigation happens. Instead, the chip provides more packets with each interrupt.
