logoalt Hacker News

tcbawoyesterday at 1:29 PM1 replyview on HN

It would not surprise me at all if the sequencing step was done via FPGA processing many network inputs at line rate with a shared monotonic clock. This would give it some amount of parallelism.


Replies

cgioyesterday at 1:33 PM

good point, sequencing is very minimal, therefore some parallelism is feasible that way, but the pipeline is not that deep, at least ideally. Of course if people are chasing nano-seconds, it may make sense.