Thanks, Guys...
In the Intel paper, if I am not mistaken, the performance is benchmarked with 2
flows. Wonder what the performance would be like with 100s or 1000s of flows.
Seems more realistic in an SP environment
Ravi
On Friday, June 26, 2015 8:03 AM, "Traynor, Kevin"
<[email protected]> wrote:
> From: discuss [mailto:[email protected]] On Behalf Of Ngo Doan
> Lap
> Sent: Thursday, June 25, 2015 5:51 PM
> To: Ravi Ramaswamy
> Cc: [email protected]
> Subject: Re: [ovs-discuss] OVS Performance
>
> Hi,
> Just for 1), please refer to test case 3,7,8,9
> here https://01.org/sites/default/files/page/intel_onp_server_release_1.3.1_b
> enchmark_test_report.pdf
There is some information that is a bit more up to date wrt OVS with DPDK if you
look through this thread
http://openvswitch.org/pipermail/dev/2015-May/055509.html
>
>
> On Thu, Jun 25, 2015 at 9:36 AM, Ravi Ramaswamy <[email protected]> wrote:
> Hello - Apologies. Newbie to OVS and this is my first posting. I am trying
> to get a high level understanding of the following. Any pointers is greatly
> appreciated
>
> * IP forwarding performance and dependence on packet size, flow distribution,
> etc
> * MPLS support on OVS
> * Multi-tenant support (I am thinking VRFs / MPLS/VPNs)
>
> Thanks
>
> Ravi
>
> _______________________________________________
> discuss mailing list
> [email protected]
> http://openvswitch.org/mailman/listinfo/discuss
>
>
>
>
> --
> Thanks and Best Regards,
> Ngo Doan Lap
> Mobile: 0977.833.757
_______________________________________________
discuss mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/discuss
