Open Stack

On 04/05/16 19:00, "Edward Leafe" <ed at leafe.com> wrote:

>On May 3, 2016, at 7:05 PM, Mark Doffman <mdoffman at linux.vnet.ibm.com> wrote:
>
>> This thread has been a depressing read.
>> 
>> I understand that the content is supposed to be distributed databases but for me it has become an inquisition of cellsV2.
>
>Thanks for bringing this up, and I feel a lot of the responsibility for this direction. To make how I see things clearer, I wrote a follow-up blog post [0], but for those who aren’t inclined to read it, I think that Cells V2 is a great idea and could be very helpful for many deployments. My only concern was the choice of fragmenting the data. I would hope that any further discussion focuses on that.
>
>[0] http://blog.leafe.com/index.php/2016/05/04/mea-culpa-and-clarification/
>
>
>-- Ed Leafe
>
>

From the perspective of an operator who is running more than 30 cells with over 6,000 hypervisors, we are strongly supporting the Cells V2 work, including investing significant effort with the CERN collaboration with BARC in Mumbai. The roadmap seems very concrete which is key for us as a production cloud at scale.  In particular, inclusion of blocking tests in the gate will be key to address current functional limitations such as flavors, server groups and security groups with cells v1.

OpenStack provides lots of opportunities for investigating alternative approaches but we do need to be careful to not end up with a Horizon Effect (https://en.wikipedia.org/wiki/Horizon_effect) where potential new solutions cause a postponing of solutions for solid production solution at scale which is needed by an increasingly large number of deployments.

Tim
>
>
>
>
>__________________________________________________________________________
>OpenStack Development Mailing List (not for usage questions)
>Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev