[Openstack] What is the best network architecture for Openstack

2014-09-18 Thread yasith tharindu
We have four servers and one SAS data store for the openstack deployment.
All servers have SAS interfaces. We are going to put one controller node
and 3 compute nodes. We need to run the system with live migration enabled.

So the option is mount the SAS partition to the controller and share it
with the computes through NFS. But by doing this, only controller SAS
interface will be used and probably network floods. Compute nodes SAS will
be unusable.


Is there are other options available to use the Compute nodes SAS
interfaces too but live migration needed also enabled.



-- 
Thanks..
Regards...

Blog: http://www.yasith.info
Twitter : http://twitter.com/yasithnd
LinkedIn : http://www.linkedin.com/in/yasithnd
GPG Key ID : *57CEE66E*
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack


Re: [Openstack] What is the best network architecture for Openstack

2014-09-18 Thread yasith tharindu
Thanks for the reply Robert;

Does GFS is stable now ? Are there Openstack production setups runs with
GFS?

On Thu, Sep 18, 2014 at 12:17 PM, Robert van Leeuwen 
robert.vanleeu...@spilgames.com wrote:

  We have four servers and one SAS data store for the openstack
 deployment. All servers have SAS
  interfaces. We are going to put one controller node and 3 compute nodes.
 We need to run the system with
  live migration enabled.
 
  Is there are other options available to use the Compute nodes SAS
 interfaces too but live migration needed also
  enabled.

 The only way to directly access the same data on the SAS data store would
 be by running a clustered filesystem. (e.g. GFS)
 Clustered filesystems are quite a bit more complex then regular
 filesystems so I'm not sure if that would be the preferred route though.


 Cheers,
 Robert van Leeuwen









-- 
Thanks..
Regards...

Blog: http://www.yasith.info
Twitter : http://twitter.com/yasithnd
LinkedIn : http://www.linkedin.com/in/yasithnd
GPG Key ID : *57CEE66E*
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack


Re: [Openstack] What is the best network architecture for Openstack

2014-09-18 Thread yasith tharindu
Thanks for the information Mike, I would initially test with GlusterFS.

On Thu, Sep 18, 2014 at 8:19 PM, Mike Smith mism...@overstock.com wrote:

  We run GFS2 for our shared instances volume and it has worked very well
 for us.  We leverage CLVM and fiber connected SAN LUNs on the backend in
 order to do that.  It's great for live migrations, etc.

  I'm told that GFS2 performance degrades significantly once you mount the
 filesystem on more than about 16 nodes.   We are planning on moving to
 GlusterFS or Ceph in the future for a couple of reasons:

  - Ceph and Cluster scale-out more linearly
 - We want to use more commodity-type hardware in remote data centers
 - We don't any want cluster/quorum related issues to take down the shared
 storage



  Mike Smith
 Principal Engineer, Website Systems
 Overstock.com


   --
 *From:* yasith tharindu [yasithu...@gmail.com]
 *Sent:* Thursday, September 18, 2014 4:06 AM
 *To:* Robert van Leeuwen
 *Cc:* openstack@lists.openstack.org
 *Subject:* Re: [Openstack] What is the best network architecture for
 Openstack

   Thanks for the reply Robert;

  Does GFS is stable now ? Are there Openstack production setups runs with
 GFS?

 On Thu, Sep 18, 2014 at 12:17 PM, Robert van Leeuwen 
 robert.vanleeu...@spilgames.com wrote:

  We have four servers and one SAS data store for the openstack
 deployment. All servers have SAS
  interfaces. We are going to put one controller node and 3 compute
 nodes. We need to run the system with
  live migration enabled.
 
  Is there are other options available to use the Compute nodes SAS
 interfaces too but live migration needed also
  enabled.

 The only way to directly access the same data on the SAS data store would
 be by running a clustered filesystem. (e.g. GFS)
 Clustered filesystems are quite a bit more complex then regular
 filesystems so I'm not sure if that would be the preferred route though.


 Cheers,
 Robert van Leeuwen









  --
 Thanks..
 Regards...

 Blog: http://www.yasith.info
 Twitter : http://twitter.com/yasithnd
 LinkedIn : http://www.linkedin.com/in/yasithnd
 GPG Key ID : *57CEE66E*

 --

 CONFIDENTIALITY NOTICE: This message is intended only for the use and
 review of the individual or entity to which it is addressed and may contain
 information that is privileged and confidential. If the reader of this
 message is not the intended recipient, or the employee or agent responsible
 for delivering the message solely to the intended recipient, you are hereby
 notified that any dissemination, distribution or copying of this
 communication is strictly prohibited. If you have received this
 communication in error, please notify sender immediately by telephone or
 return email. Thank you.




-- 
Thanks..
Regards...

Blog: http://www.yasith.info
Twitter : http://twitter.com/yasithnd
LinkedIn : http://www.linkedin.com/in/yasithnd
GPG Key ID : *57CEE66E*
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack