It was also linked to from HN: https://news.ycombinator.com/item?id=8167126
Both have some pretty good sources (and some not so good sources too)
Shipped in FreeBSD by default, developed for FreeBSD first -- and that's just the bleeding edge side of things.
Time for a epoll for kqueue swap, and make this performance debate go away for both, Linux and FreeBSD once and for all. No reason for this pissing contest.
(As in, the stuff that facilitates registered I/O is based on concepts that can be traced back to VMS, released in 1977. Namely, the Irp, plus, a kernel designed around waitable events, not runnable processes.)
epoll requires more syscalls to do the same stuff.
that's not responsible for the differences speed, but at this level syscalls are a meaningful expense.
I remember these claims being made in the late 90s, and perhaps they were true back then, but it's been 15 years, and I would be surprised if Linux hasn't caught up by virtue of its faster development pace, greater mindshare, and increased corporate/datacenter usage.
So, in all seriousness: what recent, well argued essays/papers can you refer me to so I can understand the claim that BSD networking is still better than Linux in 2015?