IPv6 throughput only half of IPv4 on ERL

[Adrian Chadd]

everything's bigger in ipv6?

Well, try kldload hwpmc and see if pmcstat works for you. I haven't
tried it on the Octeon hardware. That'll be useful for chasing down
why it's chewing CPU.


-adrian


On 8 February 2015 at 05:00, Wolfgang Zenker <wolfgang at lyxys.ka.sub.org> wrote:
> Hi,
>
> I just started to experiment with FreeBSD MIPS, using an Edge Router Lite
> as testmachine, running a recent 10.1-STABLE.
>
> To test network throughput, I enabled the discard port in inetd and transfered
> a large file there, via both IPv4 and IPv6.
>
> Results:
> $ dd if=vc.tar bs=1m|nc -4 pomona.lyx 9
> 552+1 records in
> 552+1 records out
> 579002368 bytes transferred in 21.193134 secs (27320281 bytes/sec)
> $ dd if=vc.tar bs=1m|nc -6 pomona.lyx 9
> 552+1 records in
> 552+1 records out
> 579002368 bytes transferred in 38.638252 secs (14985211 bytes/sec)
>
> Testing the same against an Atom-based machine on the same switch
> took 8.725300 secs (v4) and 10.364100 secs (v6), so the difference
> between both protocol stacks is much smaller on the Atom target.
>
> Any ideas what is the cause of this huge difference on the MIPS
> platform? Just curious ...
>
> Wolfgang
> _______________________________________________
> freebsd-mips at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-mips
> To unsubscribe, send any mail to "freebsd-mips-unsubscribe at freebsd.org"