Dear Kevin,
I upgrade to 7.0RC3 but still the same. 418Mbit is the roof.
older fbsd's are faster than newer.
How are the nic's connected to the cpu?
lspci -v
V7 is not (in my experience) slower than V4, v5, or v6.
v6 is at least slower than v4.
http://www.tancsa.com/blast.html
(look at the table at the end)
I have run a lot
of tests at speeds MUCH higher than 1Gb. With 10Gb cards, I can sustain
transfer rates of over 9Gbps (assuming low RTT and suitable
hardware). 1Gbps is not even a challenge...even over a 100 ms. RTT.
You can route 9Gbps - or only source or sink 9gbps?
What packet size?
Whats the maximum pps (with 64byte packets)? (Thats the real interesting
value, not mbps)
I have a 1.2Ghz Pentium-M appliance, with 4x 32bit, 33MHz pci intel e1000
cards.
With maximum tuning I can "route" ~400mbps with big packets and ~80mbps
with 64byte packets.
around 100kpps, whats not bad for a pci architecture.
To reach higher bandwiths, better busses are needed.
pci-express cards are currently the best choice.
one dedicated pci-express lane (1.25gbps) has more bandwith than a whole
32bit, 33mhz pci-bus.
Note that high throughput may require some tuning. Transmit and receive
windows need to be rather large if the RTT is very long at all. (See
"bandwidth-delay product" in Stevens or some other TCP reference.)
I'm not shure if he's using the nic for a server or for a router?
Kind regards,
Ingo Flaschberger
_______________________________________________
freebsd-performance@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
