Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

How exactly are Distributed File Systems used in cloud environment ?

More precisely:

  • Are live VMs images (or their filesystem) usually located in the DFS ?
  • Are VMs usually used to run the backbone (actual code) of DFS structure ?

Precise example citing DFS (ceph, Gluster, GFS, GPFS, Lustre) or cloud environment (Openstack , CloudStack, ...) would be appreciated, even if I'm more interested by ceph on OpenStack for now.

I couldn't find real life example, and all this is very new for me and very confusing. I know pretty well LXC and linux administration (except from cloud solution). I went through ceph quick starting guide, and requirements. Read papers on Openstack and viewed presentation videos.

juju has charms to setup ceph, but juju can administer LXC on local machine as well as instances on cloud services from amazon or openstack and these are often VMs. But aren't these cloud services needing a DFS for their own services ? This is where it becomes quite confusing for me. And as I guess that the answer might be different for different solutions, I would like some precise example citing DFS or cloud environment involved with the given answer.

share|improve this question

closed as not constructive by Sven, Greg Askew, Ward, Magellan, mgorven Nov 17 '12 at 20:12

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.If this question can be reworded to fit the rules in the help center, please edit the question.

if these question are improper, please take time to explain why. – vaab Nov 17 '12 at 14:21
Too broad. Questions on ServerFault need to be answerable in the specific. And the answer to this is dependent on each provider's implementation. – Magellan Nov 17 '12 at 19:40
Plus you are asking us to do research for you, which makes this sound like a homework question. – Magellan Nov 17 '12 at 19:50
Should I edit my answer to be more precise ? I wanted to have a broad view of actual usage of DFS in cloud environment, but I can be more precise as I'm more interested in Ceph and OpenStack. And this is definitively not an homework question ! – vaab Nov 17 '12 at 20:39
If that is the question you're asking, StackExchange may not be the right place to ask it. – Magellan Nov 17 '12 at 20:58

In OpenStack, the one place where a distributed file system comes into play is if you want to support shared storage based live migration. If this is the case, then the files that contain the disks for the running virtual machine instances must reside on a shared filesystem, although OpenStack is agnostic to what solution you use (e.g., NFS, Gluster, Ceph ).

Also, I'm not sure if the Ceph FS (Ceph's distributed file system implementation) is considered production-ready yet.

share|improve this answer
slight correction: live migration needs that the block devices reside on shared storage, not necessarily a shared filesystem. It could be a shared block device, like iSCSI, nbd, AoE, or ceph's rbd. That last option is the 'Rados Block Device', which is directly supported by KVM and recent Linux kernels. It also has been considered production-ready for some time (unlike cephfs, which only recently is getting some real deployments and solid support). – Javier Aug 7 '13 at 13:57

Not the answer you're looking for? Browse other questions tagged or ask your own question.