Hi Foucault,
Please see
http://www.opencontrail.org/evaluating-opencontrail-virtual-router-performance/
We get line rate (~9Gbps) unidirectional with 1500 MTU on the physical
interface.
The results can vary depending on which NUMA node the VM and the vhost-net
process
(associated with the VM) run. Please try pinning them to the same NUMA node.
Raja
From: Dev <[email protected]> on behalf of Foucault de
Bonneval <[email protected]>
Date: Monday, June 27, 2016 at 9:54 AM
To: "[email protected]" <[email protected]>
Subject: Re: [opencontrail-dev] Opencontrail network throughput performance
Hi Jakub,
MTU is at 9000 on the physical, the bond ans the vhost.
MTU inside the VM is 1500.
Foucault
2016-06-27 18:48 GMT+02:00 Jakub Pavlik
<[email protected]<mailto:[email protected]>>:
HI Foucault,
do you have MTU 9000 everywhere? Even tap interface outside and inside of VM?
Jakub
On 27.6.2016 18:35, Foucault de Bonneval wrote:
Hi there,
We are actually running new benchmarks on Opencontrail 2.21.x version.
The maximum throughput we can get is :
- ~6Gb/s in "half duplex" from VM A -> VM B or VM B -> VM A
- ~3.5Gb/s in "full duplex" in a VM A <-> VM B pattern
Tests are run with Iperf on Ubuntu 14.04, computes are also on Ubuntu 14.04,
kernel 3.13.0-83-generic
Does anyone have run similar tests and may have results to share ?
If anyone have dig on optimizations on a system level that could improve those
results, I'd love some informations.
Thanks,
Foucault
CPU : Intel(R) Xeon(R) CPU E5-2670 v3 @ 2.30GHz (48 cores)
# lspci
01:00.0 Ethernet controller: Intel Corporation 82599ES 10-Gigabit SFI/SFP+
Network Connection (rev 01)
01:00.1 Ethernet controller: Intel Corporation 82599ES 10-Gigabit SFI/SFP+
Network Connection (rev 01)
# ethtool -k vhost0
Features for vhost0:
rx-checksumming: off [fixed]
tx-checksumming: on
tx-checksum-ipv4: on [fixed]
tx-checksum-ip-generic: off [fixed]
tx-checksum-ipv6: off [fixed]
tx-checksum-fcoe-crc: off [fixed]
tx-checksum-sctp: off [fixed]
scatter-gather: on
tx-scatter-gather: on [fixed]
tx-scatter-gather-fraglist: off [fixed]
tcp-segmentation-offload: on
tx-tcp-segmentation: on [fixed]
tx-tcp-ecn-segmentation: off [fixed]
tx-tcp6-segmentation: off [fixed]
udp-fragmentation-offload: off [fixed]
generic-segmentation-offload: on
generic-receive-offload: on
large-receive-offload: off [fixed]
rx-vlan-offload: off [fixed]
tx-vlan-offload: off [fixed]
ntuple-filters: off [fixed]
receive-hashing: off [fixed]
highdma: off [fixed]
rx-vlan-filter: off [fixed]
vlan-challenged: off [fixed]
tx-lockless: off [fixed]
netns-local: off [fixed]
tx-gso-robust: off [fixed]
tx-fcoe-segmentation: off [fixed]
tx-gre-segmentation: off [fixed]
tx-ipip-segmentation: off [fixed]
tx-sit-segmentation: off [fixed]
tx-udp_tnl-segmentation: off [fixed]
tx-mpls-segmentation: off [fixed]
fcoe-mtu: off [fixed]
tx-nocache-copy: off
loopback: off [fixed]
rx-fcs: off [fixed]
rx-all: off [fixed]
tx-vlan-stag-hw-insert: off [fixed]
rx-vlan-stag-hw-parse: off [fixed]
rx-vlan-stag-filter: off [fixed]
l2-fwd-offload: off [fixed]
_______________________________________________
Dev mailing list
[email protected]<mailto:[email protected]>
http://lists.opencontrail.org/mailman/listinfo/dev_lists.opencontrail.org
--
Jakub Pavlik
CTO
[tcp ◕ cloud]
+420 602 177 027<tel:%2B420%20602%20177%20027>
[email protected]<mailto:[email protected]>
tcp cloud a.s.
Thamova 16
186 00 Praha 8 - Karlin
Czech republic
http://tcpcloud.eu
http://opentcpcloud.org
_______________________________________________
Dev mailing list
[email protected]
http://lists.opencontrail.org/mailman/listinfo/dev_lists.opencontrail.org
