Originally posted at https://vinfrastructure.it/2016/01/rubrik-reinvent-the-backup/
Some months Rubrik announced the release of version 2.0 of Rubrik Converged Data Management to deliver complete data protection and rich data services at a global scale (see also my previous post).
The idea is quite simple but disruptive: reinvent the backup models in order to eliminate backup software by integrating data protection, instant recovery, and DevOps infrastructure into a single scale-out product!
Data protection is usually something complex, or at least with several components:
Rubrik can remove all of them:
An simple move to converged, scale-out and appliance based architecture:
The idea is similar to Cohesity approach (for example) but Rubrik stats with the data protection in mind as a first scope of their solution, so several features (in this area) are better developed (yet). It’s interesting see also people from Symantec, Data Domain, Veritas in the investors list.
During the last IT Press Tour #17 I’ve got the opportunity to learn more about this solution, directly the board of Rubrik: Bipul Sinha (CEO), Arvind Jain, Arvind Nithrakashyap. For the technical presentation there was also Chris Wahl.
As written the solution is (physical) appliance based, using the model of Data Domain (that has change the tradition backup infrastructure), but using a “Web-Scale” model typical of hyper-converged product, with a grow as you go approach in 2U increments. And with a really easy setup and configuration: it can be up and running in 15 minutes!
One of the best quote is: “it looks simple because we make it simple”, Bipul Sinha. It’s really simple to use and with a fast interface.
The r300 Series is a Converged Data Management appliance that provides automated backup, instant recovery, unlimited replication, and data archival at infinite scale.
The Rubrik cluster should start from three appliances, each of them can have 12x4TB HDD or 12x8TB HDD, plus 4x400GB SSD, and can grow as you need. Deduplication is cluster-wide to improve space usage.
Actually it support VMware vSphere 5.1, 5.5 or 6.0 and you have only to point to a vCenter Server and use policy based rules (called SLA Domains) to protect your VMs. At this time only VMware platform is supported, but the entire architecture is platform agnostic, so could be possible see new virtualization platform supported in the future.
In order to provide application consistency, Rubrik push a VSS provider inside each VMs.
As show in the previous picture it’s possible replicate or simple increase the retention with an external object storage: S3 is supported, but for on-prem infrastructure (if needed) could be used S3 (for example with Scality) or Swift. Also NFS could be used for on-prem long term retention.
For site replication more Rubrik appliance could be used, both for DR purpose or Dev/Test cases.
And what about the data services, like the ability to have application-level recovery? Actually this is not included or integrated and it’s far from other products (like Veeam). Rubrik has a partnership with Kroll on Track to provide this services, but I think that simplicity and convergence mean also have those kind of services inside the solution itself and I hope to see something in the next releases.