Open Stack

Got it. By RPC I was referring to RabbitMQ in particular. That's also the rationale that Rackspace presented at the Portland summit.

Subbu

On Oct 3, 2013, at 11:42 AM, Chris Behrens <cbehrens at codestud.com> wrote:

> 
> On Oct 3, 2013, at 10:23 AM, Subbu Allamaraju <subbu at subbu.org> wrote:
> 
>> Hi Tim,
>> 
>> Can you comment on scalability more? Are you referring to just the RPC layer in the control plane?
> 
> Not just RPC, but RPC is a big one.  Cells gives the ability to split up and distribute work.  If you divide hypervisors into different cells, in each cell you have less connections to rabbit, less work for the host scheduler to find a suitable hypervisor, etc.  You also mentioned 'failure domains' and that can be a good reason to use cells as well.  I felt that the multi-DC or multi-continent scenario where you want your nova-api endpoints to see ALL instances (as opposed to multi-region with keystone) was a good use case for cells.  You should want services to talk to each other with low latency.  Each cell has its own rabbit and DB that you can keep 'close by'.  Sure, you can cluster/HA/replicate rabbit and 1 big DB, but those technologies fail and when the fail, they tend to fail badly.  Cells can isolate issues to a subset of your cloud.
> 
> - Chris
> 
> 
>> 
>> Subbu
>> 
>> 
>> On Oct 3, 2013, at 8:53 AM, Tim Bell <Tim.Bell at cern.ch> wrote:
>> 
>>>  
>>> At CERN, we’re running cells for scalability. When you go over 1000 hypervisors or so, the general recommendation is to be in a cells configuration.
>>>  
>>> Cells are quite complex and the full functionality is not there yet so some parts will need to wait for Havana.
>>>  
>>> Tim
>>>  
>>> From: Dmitry Ukov [mailto:dukov at mirantis.com] 
>>> Sent: 03 October 2013 16:38
>>> To: openstack at lists.openstack.org
>>> Subject: [Openstack] Cells use cases
>>>  
>>> Hello all,
>>> I've really interested in  cells but unfortunately i can't find any useful use cases of them.
>>> For instance I have 4 DCs and I need single entry point for them. In this case cells are  a bit complicated  solution. It's better to use multiple regions in keystone instead
>>>  
>>> The only one good reason for cells, which I've found, is to organize so-called failure domains, i.e. scheduling on another DCs in case of failures.
>>>  
>>> Does anyone have different use cases or vision on cells usage?
>>> Thanks in advance.
>>>  
>>> -- 
>>> Kind regards
>>> Dmitry Ukov
>>> IT Engineer
>>> Mirantis, Inc.
>>>  
>>> _______________________________________________
>>> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>> Post to     : openstack at lists.openstack.org
>>> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> _______________________________________________
>> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> Post to     : openstack at lists.openstack.org
>> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>