Yes: as the document there notes, the VDS driver uses provider networks (which 
in Neutron are created by the admin rather than the tenant and basically map 
neutron networks to VLANs in the physical datacenter network) and therefore 
punt L3 operations to the datacenter network.

At Your Service,

Mark T. Voelker



> On Sep 24, 2015, at 12:14 PM, Federico Michele Facca 
> <[email protected]> wrote:
> 
> As early mentioned this week, Mark's option is a valid alternative (separate 
> regions). @ Mark, I wonder if this excpert from your pointer is still valid:
> 
> "VDS-based networking has limitations, including the inability of tenants to 
> create their own private L2 networks, and the inability to deliver L3 and 
> higher networking services such as virtual routers, security groups, and 
> floating IPs. 
> If such features are important for your VMware Integrated OpenStack 
> deployment, consider using NSX-V for Neutron networking."
> 
> In case I think people should be aware of this before going for this option.
> 
> Br,
> Federico
> 
> --
> Future Internet is closer than you think!
> http://www.fiware.org
> 
> Official Mirantis partner for OpenStack Training
> https://www.create-net.org/community/openstack-training
> 
> -- 
> Dr. Federico M. Facca
> 
> CREATE-NET
> Via alla Cascata 56/D
> 38123 Povo Trento (Italy)
> 
> P  +39 0461 312471
> M +39 334 6049758
> E  [email protected]
> T @chicco785
> W  www.create-net.org
> 
> On Thu, Sep 24, 2015 at 5:58 PM, Mark Voelker <[email protected]> wrote:
> Miko,
> 
> There’s a VDS driver for Neutron in addition to the NSX one that you might 
> want to have a look at if you haven’t already.  Some notes on that here 
> (these are from the admin guides for the VMware Integrated OpenStack 
> distribution as it does a pretty good job of describing the requirements, 
> architecture, and limitations of the VDS driver, but the driver itself is 
> open source and can be used by anyone):
> 
> http://pubs.vmware.com/integrated-openstack-2/index.jsp#com.vmware.openstack.install.doc/GUID-98D92A0C-E568-4380-A9AA-7AE9A59E360E.html
> 
> VDS obviously won’t work for the KVM compute hosts though.  A fairly 
> widely-used tactic if you don’t want to use an SDN or overlay based solution 
> for spanning multiple hypervisors is to separate the hypervisor hosts into 
> separate regions (which have distinct API endpoints and can therefore have 
> services configured completely differently if you so desire).  If you’re not 
> familiar with the concept of regions, have a look here:
> 
> http://docs.openstack.org/openstack-ops/content/scaling.html
> 
> Note that if you go that route users need to target one region or the other 
> when making API calls since they are distinct API endpoints.
> 
> At Your Service,
> 
> Mark T. Voelker
> 
> 
> 
> > On Sep 24, 2015, at 5:01 AM, Miko Bello <[email protected]> wrote:
> >
> > Hi Folks,
> > i would like to know if anybody has tried to implement a neutron solution 
> > with vsphere without using NSX-like solutions.
> > I mean, my lab environment is composed of :
> >
> >  - 1 network node ( neutron )
> >  - 1 controller node
> >  - 2 compute node ( kvm)
> >  - 1 compute node linked to a cluster vsphere 6.0
> >
> > i tried, without success ;( to deploy an instance on vsphere node with a 
> > network configuration of type VLAN; ( obviously, in this configuration,i 
> > have not the ambition of a ovs solution type :)  )
> > so my questions are: it's' possible to implement a network neutron solution 
> > of type VLAN ? if yes how can i do?
> > Thanks in advantage.
> > Miko Bello
> >
> >
> >
> >
> >
> >
> >
> > _______________________________________________
> > OpenStack-operators mailing list
> > [email protected]
> > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> 
> _______________________________________________
> OpenStack-operators mailing list
> [email protected]
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> 

_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Reply via email to