[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] About ping latency in SR-IOV


  • To: "Rose, Gregory V" <gregory.v.rose@xxxxxxxxx>, "Dong, Eddie" <eddie.dong@xxxxxxxxx>
  • From: XiaYubin <xiayubin@xxxxxxxxx>
  • Date: Fri, 29 Oct 2010 16:31:09 +0800
  • Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
  • Delivery-date: Fri, 29 Oct 2010 01:32:26 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:from:date:message-id:subject:to:cc:content-type; b=jQKJaGCrHyE1lTcl1yF4uD+cMn9dgrOhPW/Snp4QCPGJLvHyU367YTdLk4BgzGEbxI E9BpYh2sMo0OGL5Uiem1q9/pKHSrF4+NW1Paazaj8sxHfgOG3y8knk06Wif7ddD3L4g0 VoPY4b6xHJ2+WOP3YEjS9NtRNqc1qfmER95ko=
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>

Hi, Greg, Eddie,

I found that in SR-IOV + HVM environment, ping latency depends on ping interval.

The NIC is Intel 82576 (1000Mbps), server uses xen-4.0.1, domain-0 is
kernel-2.6.32
and HVM is kernel-2.6.31. The server and client are connected through
a 1000Mbps switch.

The data is as following:

# ping -i 1 192.168.0.21

PING 192.168.0.21 (192.168.0.21) 56(84) bytes of data.
64 bytes from 192.168.0.21: icmp_seq=1 ttl=64 time=0.117 ms
64 bytes from 192.168.0.21: icmp_seq=2 ttl=64 time=0.100 ms
64 bytes from 192.168.0.21: icmp_seq=3 ttl=64 time=0.102 ms
64 bytes from 192.168.0.21: icmp_seq=4 ttl=64 time=0.100 ms
64 bytes from 192.168.0.21: icmp_seq=5 ttl=64 time=0.105 ms

--- 192.168.0.21 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 3998ms
rtt min/avg/max/mdev = 0.100/0.104/0.117/0.014 ms

# ping -i 0 192.168.0.21

PING 192.168.0.21 (192.168.0.21) 56(84) bytes of data.
64 bytes from 192.168.0.21: icmp_seq=1 ttl=64 time=0.114 ms
64 bytes from 192.168.0.21: icmp_seq=2 ttl=64 time=0.418 ms
64 bytes from 192.168.0.21: icmp_seq=3 ttl=64 time=0.455 ms
64 bytes from 192.168.0.21: icmp_seq=4 ttl=64 time=0.456 ms
64 bytes from 192.168.0.21: icmp_seq=5 ttl=64 time=0.455 ms

--- 192.168.0.21 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 1ms
rtt min/avg/max/mdev = 0.114/0.379/0.456/0.135 ms, ipg/ewma 0.400/0.252 ms

The ping latency of PF (in domain-0) is about 0.08ms in both tests.
Is this caused by something related to interrupt throttle configuration?
If so, can we adjust its threshold?

Thanks!

--
Yubin

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.