Use cases for Red Hat storage

Latest response

The Red Hat storage product centers around the glusterfs clustered file system. How do you know Red Hat storage is right for your needs? Here is a list of popular use cases:

 

  • Digital multi-media (video, audio, pictures) serving
  • Content delivery networks
  • Large scale unstructured file storage
  • Near-line storage
  • Backup and restore
  • File sharing
  • Select High Performance Computing (HPC) usage

Please comment about your own use cases for glusterfs and Red Hat Storage. Or ask questions about how well-suited Red Hat Storage might be for your environment.

Responses

Hi

 

I would like to understand a little more about how Red Hat Storage might integrate with the RHEV product and how we might use it to provide a cross datacentre DR (active - active) solution. 

 

Thanks

Hi Richard,

 

The following might be useful - it was a presentation on the roadmap of Red Hat Storage presented at Red Hat Summit 2012 last month:

 

http://rhsummit.files.wordpress.com/2012/03/williams_storage_roadmap_futures.pdf - check out slide 11.

 

The full list of presentations posted so far can be found at http://www.redhat.com/summit/2012/presentations/

 

Hope this helps!

 

Andrius.

A customer of mine would liked to have used it for archival storage, but it (still) lacks 2 needed features:

 

1. Versioning (essentially file level snapshotting)

2. Bit-rot detection (and restoration)

 

I am already in the (GLusterFS) mailinglists and talking with my local RedHat Solution Architect go get this into the product ASAP.

1. Versioning

 

This is being evaluated and scoped for enhancement in a future RHS release.

 

2. Automatic bit-rot detection and repair

 

We've discussed this before, but it's hard as long as we depend on local filesystems (which is for good reasons) to store the actual bits and they either lack such features or don't expose them through an API. Note that RAID (hardware and software) can do some of this without our involvement.  If we're using AFR, we could also detect copies of files with unrepairable errors at that level, and use our own self-heal to regenerate it from another full copy.

 

For full details, please follow-up with your SA.

 

Hope this helps!

 

Andrius.

 

I'm currently using RHS for scale-out nas over samba/ctdb/rr-dns (48Tb of jeps images, and other production data for our digital photo lab.)

...And Virtual Machine Images for RHEV. We currently do that with nfs/cdtb/rr-dns, but I'm really hanging out for the new versions of RHEV and RHS that will do this with the native glusterfs....