back to article Hyper-scaling multi-structured data? Let's count the ways

Hyperscaling storage for unstructured data, file and object silos is conceptually straightforward. You buy yourself commodity hardware and get parallel filesystem software or object software. Hyperscale block storage generally means buying a monster SAN. But what happens if you need to have hyperscale storage across block, file …

  1. iOS6 user

    Anyone knows what is the status cepth vs gluster?

    How things are going on those to horizontally scaled file server?

    Which one and why i more promising and/or what in the future in case one of those solutions something may be harder to implement/provide?

    1. CephDude

      Ceph is the future

      My experience is only with Ceph, I've never actually ran anything in production with Gluster.

      We have about 10 PB of Ceph in production now and we're doubling that to about 21 PB by end of this year. Ceph is incredibly stable for object and block (we use it exclusively with OpenStack).

      Ceph is not yet there for file, so if you need stable scale-out storage for file only Gluster is probably a safer bet. Inktank/RedHat have been running CephFS internally for over a year and a half, but they do not yet have a date for when they'll start supporting it. If you are a small shop, you should definitely give it a try.

      There are plenty of ways to setup Ceph, if you're familiar with Ansible, have a look at this playbook:

      You can spin up something with Vagrant to play with.

      Hope this helps.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like

Biting the hand that feeds IT © 1998–2022