> > I launched ovs-dpdk with a single GB of memory on the first numa node
> > (instructions seem to indicate memory should only be allocated to the
> > first numa node... not sure why). When I tried increasing the number
> > of RX queues (ovs-vsctl set Open_vSwitch .
> > other_config:n-dpdk-rxqs=2), performance got significantly worse. I
> > was wondering what is the proper way to scale up the performance of
> > ovs-dpdk when scaling up the number of VMs that each are using
> significant network bandwidth?
>
> Probably the best way to scale is to add more cores with pmd-cpu-mask.
> You won't get much by adding multiple rxqs unless you do this, as the same
> pmd (core) will be polling all the rxqs.
>
I think I found an issue:
ovs-appctl dpif-netdev/pmd-stats-show
main thread:
emc hits:0
megaflow hits:0
miss:0
lost:0
polling cycles:2209488 (100.00%)
processing cycles:0 (0.00%)
pmd thread numa_id 0 core_id 0:
emc hits:50
megaflow hits:6288906
miss:1
lost:0
polling cycles:4860224526 (8.11%)
processing cycles:55082803644 (91.89%)
avg cycles per packet: 9531.47 (59943028170/6288957)
avg processing cycles per packet: 8758.65 (55082803644/6288957)
pmd thread numa_id 0 core_id 8:
emc hits:0
megaflow hits:0
miss:1
lost:0
polling cycles:37034036502 (100.00%)
processing cycles:76503 (0.00%)
avg cycles per packet: 37034113005.00 (37034113005/1)
avg processing cycles per packet: 76503.00 (76503/1)
If you notice, both pmd threads have a numa_id of 0.
Also launching the ovs-vswitchd, the output shows:
...
2015-09-10T21:24:51Z|00033|dpif_netdev|INFO|Created 2 pmd threads on numa node 0
...
I thought it was supposed to create a thread on each numa node?
_______________________________________________
discuss mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/discuss
