Slashdot Mirror
986MB/s With BSD And Gigabit Ethernet
WasterDave wrote in with this link to information about zero copy sockets on FreeBSD. Some hunting turned up more detailed information about NetBSD and Gigabit networking. Pointers to similar information for OpenBSD are appreciated...
The units in this case would be in megabits per second, not megabytes per second. 986MB/s would be 7888Mbps, or 7.888Gbps.
-------
FreeBSD: The power to serve.
From the linux-kernel list:
David Miller: "make no mistake, for transmit we will at some point have a zero copy scheme available. And you can be certain that when it does happen, the end to end latency will not suffer like it does on other systems for the cases where zero copy makes no sense at all."
Linus Torvalds: "So don't fall into the trap of thinking that zero-copy is always obviously a win. It isn't."
Ingo Molnar: "zero-copy makes a RL difference only in a small part of those uses. I very much dislike zero-copy-maniac designs which give up just about everything to get nice bandwith numbers."
cpeterso
One important question is, what MTU have they used. If it's the 9000 byte MTU jumbo gigabit frames then these BSD numbers havent got too practical relevance (i can saturate 8 gigabit cards with TUX, ie. 900MB/sec with 9000 byte MTU). If it's the standard 1500 byte MTU then it's nicer. (hm, i just found it, they indeed used 9000 byte MTU...)
The "turbo packet support" feature has just appeared in test versions of pre-2.4 Linux in the past week or two. That's hardly "always had it," as our troll claims (as if a system that isn't even released yet could "have" anything). So it's no further along than the FreeBSD equivalent, and despite the claims of our troll, is just as likely to be a copy of the BSD feature as vice-versa (in other words, unlikely). But of course, the reality is that some other OSes have had zero-copy for quite a long time. So who is copying whom?
(You'll have to search the "linux-net" list, not the "linux-kernel" list, for a discussion of "turbo packet support" and not "turbo sockets." Not only is our troll such a Linux ignoramous that he doesn't know where to look, he doesn't even know what to call it.)
It's always a sign of an immature wanna-be when such bare claims of BSD or Linux superiority are made; it's pretty obvious why our troll has decided to remain anonymous. The fact is that Linux and BSD each help make the other better, with friendly competition among the actual developers and a free flow of information between them. With rare exceptions, the attitude between workers in the two camps is one of mutual respect and even occasionally admiration. (Far more dissing goes on within the groups themselves than between them -- it's the camp followers who make all the us-vs-them noises.)
The Linux vs. BSD lamers simply don't understand what free software development is all about; unlike commercial software, win/win situations are the rule, and not the exception.