<div dir="ltr">We also have SRIOV code ready and working at about the level you have, by the sounds - perhaps we could compare notes?<br></div><div class="gmail_extra"><br><br><div class="gmail_quote">On 26 February 2013 09:29, Irena Berezovsky <span dir="ltr"><<a href="mailto:irenab@mellanox.com" target="_blank">irenab@mellanox.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Great summary on required pieces to support.<br>
Supporting SRIOV vNIC is something that we (at Mellanox) dealing with.<br>
We currently support plugin (available at <a href="https://github.com/mellanox-openstack/mellanox-quantum-plugin" target="_blank">https://github.com/mellanox-openstack/mellanox-quantum-plugin</a>) with only Quantum Networking and Nova VIF parts.<br>
We plan to push this code Upstream soon.<br>
There is still the Scheduler part that need to be implemented for Host assignment based on available SR-IOV resources.<br>
<br>
Will be great if we can discuss complete SR-IOV support at the summit or over emails.<br>
<br>
Regards,<br>
Irena<br>
<div class="im HOEnZb"><br>
-----Original Message-----<br>
From: Isaku Yamahata [mailto:<a href="mailto:yamahata@valinux.co.jp">yamahata@valinux.co.jp</a>]<br>
Sent: Friday, February 22, 2013 5:53 PM<br>
To: OpenStack Development Mailing List<br>
Subject: Re: [openstack-dev] [quantum] SR-IOV VNIC support in quantum<br>
<br>
</div><div class="HOEnZb"><div class="h5">On Thu, Feb 21, 2013 at 07:28:41PM +0100, Ian Wells wrote:<br>
> It's certainly something we're looking at (in Cisco) at the moment. I<br>
> absolutely think it's worth having a session, though there are several<br>
> parts to the problem, and it crosses the nova-quantum divide which can<br>
> make it awkward to schedule:<br>
><br>
> - SR-IOV and directmap into the hypervisor (we're looking at Zardara<br>
> Storage's work on this with libvirt, which has never been incorporated<br>
> into the mainline but basically works fine)<br>
> - scheduling based on resources that are limited in number - that is,<br>
> scheduling to a machine that has remaining unallocated SRIOV NICs (or<br>
> whatever<br>
> else)<br>
> - how one might get an SRIOV NIC into a Quantum network<br>
<br>
Assuming that NIC has features like anti-spoofing/vlan tag insertion in NIC l2 switching logic(I suppose intel NIC has such features), Quantum networking can work with OpenFlow top-of-rack switch.<br>
<br>
Probaly the following blue print can be enhanced to support SR-IOV Generic Hardware Driver interface for Quantum Plugins <a href="https://blueprints.launchpad.net/quantum/+spec/ovsplugin-hardware-devices" target="_blank">https://blueprints.launchpad.net/quantum/+spec/ovsplugin-hardware-devices</a><br>
<br>
thanks,<br>
<br>
> And of course, not all devices that you might wish to directmap are NICs.<br>
><br>
> Migration's not something I've been giving much thought to, I admit.<br>
> --<br>
> Ian.<br>
><br>
><br>
> On 21 February 2013 15:45, Jiang, Yunhong <<a href="mailto:yunhong.jiang@intel.com">yunhong.jiang@intel.com</a>> wrote:<br>
><br>
> Hi, all<br>
> I want to know if there are any plan to support SR-IOV networking<br>
> in openstack?<br>
><br>
> SR-IOV network reduce the CPU utilization greatly and improve the<br>
> scalability and sharing capability greatly [1]. Combining with bonding<br>
> driver [2], it can also support live migration.<br>
><br>
> However, SR-IOV networking also have some issue traditionally. For<br>
> example, it's not easy to control flow from SR-IOV NIC, and not easy to<br>
> setup the tenant network. However, SDN can ease such limitation greatly,<br>
> also some private environment does not have highly control requirement and<br>
> will be helpful.<br>
><br>
> Any idea? Can this be a topic candidate in design summit?<br>
><br>
> Thanks<br>
> --jyh<br>
><br>
> [1] <a href="http://static.usenix.org/event/wiov08/tech/full_papers/dong/dong.pdf" target="_blank">http://static.usenix.org/event/wiov08/tech/full_papers/dong/dong.pdf</a><br>
> [2] <a href="http://www.kernel.org/doc/ols/2008/ols2008v2-pages-261-267.pdf" target="_blank">http://www.kernel.org/doc/ols/2008/ols2008v2-pages-261-267.pdf</a><br>
><br>
> _______________________________________________<br>
> OpenStack-dev mailing list<br>
> <a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
><br>
><br>
<br>
> _______________________________________________<br>
> OpenStack-dev mailing list<br>
> <a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
<br>
--<br>
yamahata<br>
<br>
_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</div></div></blockquote></div><br></div>