On Tue, May 13, 2014 at 10:38 PM, Vasiliy Tolstov <[email protected]> wrote: > 2014-05-14 0:09 GMT+04:00 Jesse Gross <[email protected]>: >> All of these tunnels are going to perform similarly. ARP suppression >> would need to be implemented on any of them using some kind of >> controller if that is really the issue. > > > Okay. What about performance and scalability? As i understand in gre i > can use multipatch tcp, but in case of vxlan i can't use it and need > to use for example equal cost multipath routing (i don't use bonding).
VXLAN generally has better interaction with ECMP but otherwise they are the same. I would recommend reading about VXLAN and GRE as there are quite a few articles floating around. _______________________________________________ discuss mailing list [email protected] http://openvswitch.org/mailman/listinfo/discuss
