FreeBSD problems and preliminary ways to solve
gpalmer at freebsd.org
Fri Aug 19 17:22:54 UTC 2011
On Fri, Aug 19, 2011 at 05:53:07PM +0400, Lev Serebryakov wrote:
> Hello, Pieter.
> You wrote 19 ??????? 2011 ?., 16:01:23:
> >> > 3. Kernel features for complex network solutions (netgraph, carp, ipfw).
> >> >
> >> > The niche for routers & traffic analysis is still ours. It would be
> >> > nice to take e.g. pfSense and agree with some vendor (Netgear,
> >> > D-Link, etc) to put on sale hardware with FreeBSD inside.
> >> What about 10G routing? Here are reports about full-bandwidth 10G routing
> >> on modern Intel NICs with Linux (and multi-core server), but I didn't see
> >> any such data for FreeBSD, and somebody says, that Intel drivers and
> >> network stack is not so good parallel in FreeBSD.
> > With regards to high speed packet forwarding and routing, check out this work
> > by Luigi Rizzo:
> > http://info.iet.unipi.it/~luigi/netmap/
> Yep, I know this work, but it is rather special, not out-of-box (ok,
> maybe with some changed tunables) routing solution with routing
> tables, some firewall, etc.
I think we need to refocus on what would give the most benefit. At least
as far as I can see:
- port/package management: most users would benefit
- improved virtualisation support: a lot of users would benefit
- improved driver support: a lot of users would benefit
- KMS/GEM for better graphics card support: a lot of users would benefit
Can you honestly say the same about handling line rate packet forwarding
for multiple 10G cards?
I suspect I would rank true multipath storage support above 10G routing,
More information about the freebsd-arch