On Mon, 10 Mar 2014 01:31:13 +0400
Vasiliy Tolstov <[email protected]> wrote:

> 2014-03-05 17:42 GMT+04:00 FUJITA Tomonori <[email protected]>:
>> The driver is called 'OpenFlow Agent mechanism driver', not 'Ryu
>> something'. As the name says, it's not the second version of the
>> current Ryu driver. It adopts the different design; unlike the current
>> Ryu driver (and other OpenFlow controller drivers), 'OpenFlow Agent
>> mechanism driver' has no central OpenFlow controller. Instead,
>> controllers are deployed on each compute nodes for better scalability
>> and availability.
> 
> 
> Hi. But how in this case controller from one node can create shortest
> path for another vm on different node ?

I'm not sure I understand what you mean but I guess your concern is
that there is no central managenement point with the new approach?  If
so, there is still central management point so you can cordinate
multiple controllers in your way. The new approach exploits the
existing Neutron central management point and let OpenStack developers
work on its scalability, availablity, other tricky issues about the
central management point.

> Where i can find more info about benefits of controller per compute node?

There are others who use the similar approach. The following is one
example:

http://www.slideshare.net/junparkearth?utm_campaign=profiletracking&utm_medium=sssite&utm_source=ssslideview

------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/13534_NeoTech
_______________________________________________
Ryu-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ryu-devel

Reply via email to