Network performance measurements of -current

ShouYan Mao symao at juniper.net
Fri Dec 23 01:15:56 PST 2005


So, would you try it with PCI-E Gigabit card?

Best Regards.
Shouyan

-------------------------------------------------------
I'm not the best, but I try to do better.
-------------------------------------------------------

-----Original Message-----
From: owner-freebsd-net at freebsd.org =
[mailto:owner-freebsd-net at freebsd.org] On Behalf Of Andre Oppermann
Sent: 2005=C4=EA12=D4=C223=C8=D5 2:17
To: freebsd-current at freebsd.org
Cc: freebsd-net at freebsd.org
Subject: Network performance measurements of -current

As part my funded TCP/IP optimization work I'm doing lots of =
measurements
and profiling with an Agilent N2X network tester and calibrated traffic
generator.

The following data shall serve as baseline of the current performance we
get out of FreeBSD 7-current.  More to come tomorrow though.

OS: FreeBSD 7-current as of 20051222-1600 UTC
KERNEL: Generic kernel, minus WITNESS and INVARIANTS, plus HWPMC, =
HZ=3D1000
HARDWARE: Dual Opteron 852 2.6Ghz, Tyan S2882 Mobo with AMD-8131 PCI-X =
tunnel
HARDWARE: dual Broadcom Gigabit BMC5704C PCI-X-133 ("bge")
HARDWARE: dual Intel Gigabit 82546EB PCI-X-133 ("em")

Uniprocessor kernel

 bge:
  normal forwarding bge0->bge1:   =
@64/326kpps/166us/402kpps(30%Loss)/194us
  normal forwarding bge0->bge1:   @1500/81kpps/520us
  normal forwarding bge0->disc0:  @64/1205kpps
  IP fastforwarding bge0->bge1:   =
@64/565kpps/192us/575kpps(60%Loss)/1090us
  IP fastforwarding bge0->bge1:   @1500/81kpps/730us
  IP fastforwarding bge0->disc0:  @64/1160kpps
  net.isr.direct=3D1  bge0->bge1:   =
@64/476kpps/211us/487kpps(68%Loss)/1284us
  net.isr.direct=3D1  bge0->bge1:   @1500/81kpps/760us
  net.isr.direct=3D1  bge0->disc0:  @64/1250kpps
  polling (*) bge0->bge1:       =20
@64/420kpps(9%Loss)/1385us/416kpps(72%Loss)/1600us
  polling (*) bge0->bge1:         @1500/71kpps(9%Loss)/850us
  polling (*) bge0->disc0:        @64/697kpps

 Comments: Under full load the normal processing breaks completely down
  while with IP fastforwarding it levels off but continues to forward.
  Strangely with polling it has 9% loss at all loads (even at 1% =
wirespeed).
  May be related to HZ=3D1000.

 em:
  normal forwarding em0->em1:     =
@64/372kpps/112us/396kpps(11%Loss)/131us
  normal forwarding em0->em1:     @1500/81kpps/170us
  normal forwarding em0->disc0:   @64/1130kpps
  IP fastforwarding em0->em1:     =
@64/565kpps/45us/585kpps(4%Loss)/1600us
  IP fastforwarding em0->em1:     @1500/81kpps/135us
  IP fastforwarding em0->disc0:   @64/1116kpps
  net.isr.direct=3D1  em0->em1:     later
  net.isr.direct=3D1  em0->disc0:   later
  polling (*) em0->em1:           later
  polling (*) em0->disc0:         later


(*) max_burst=3D1000, user_frac=3D0, each_burst=3D30

Sponsored by:	TCP/IP Optimization Fundraise 2005

--=20
Andre
_______________________________________________
freebsd-net at freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"


More information about the freebsd-current mailing list