Should Data Protection be Taken to a Higher Level - the OpenStack Cloud?
10 years ago data protection (primarily backup and recovery) worked at the physical server level. When server virtualization gained popularity, data protection vendors first said that they could continue to work at the host level until the end users demanded that their data protection vendors rise to a higher level and recognize the virtual infrastructure. Of course with that came some amazing features (like image level backup, change block tracking, and more).
Recently I was at OpenStack Summit 2016 in Austin, Texas where I met with the CEO of Trilio Data, David Safaii. His company is the only data protection vendor who has already taken data protection to an even higher level – to the OpenStack cloud.
This is important because companies are moving, more and more, to cloud-like solutions – whether it’s public cloud or private cloud – and as part of that “cloud-like consumption model” they expect things to be “easy”. Part of that definition of “easy” is being able to provision the virtual machines and applications that you need, when you need them, and then destroy those virtual machines, when you no longer need them. In many cases the consumer’s cloud infrastructures are built up in seconds (using automated tools), consumed, and then destroyed in seconds (again using automated tools). BUT WAIT – what if you wanted some of that data back? As part of the easy, self-service infrastructure solution surely you can push a button and recover those virtual machines, right?
Well.. not so fast… Just because you consume a private cloud or a public cloud doesn’t mean that data protection is automatically built in, or included. However, many consumers expect data protection to be automatically part of any cloud-solution. However, most service providers offering public cloud solutions or enterprises offering private clouds still use traditional data protection solutions that are virtualization-aware but not cloud-aware.
Traditional virtualization backup solutions backup at the virtual data center level, the cluster level, or the virtual machine level. When a VM running in a cloud needs to be recovered, it may be possible but it may not be automatic and it’s unlikely that it can be done through self-service.
On the other hand, Trilio Data’s solution, called TrilioVault, backs up and recovers “OpenStack Clouds”. TrilioVault is the industry’s first native backup and recovery solution for OpenStack. Developed out of Raksha – the first data-protection-as-a-service specification for OpenStack – and authored by Trilio Data’s founders, TrilioVault was designed and built from the ground up specifically for OpenStack.
The end result is that, with TrilioVault in place, consumers who are using public OpenStack powered clouds or who are using private OpenStack-powered clouds do have automatic data protection for their virtual machines and, should they need recovery, those consumers are able to use a self-service portal to recover their data, just as they would expect that they could.
An important differentiation with TrilioVault is that, unlike a traditional virtualization-backup tool that can protect a VM, TrilioVault is focused on the entire cloud environment and, as such, they protect not just a single VM but groups of VMs as well as the the compute, network, and storage configurations. With that, a group of VMs (including the applications and OS inside) can be restored at a single point in time along with the cloud configuration at that point in time. That metadata about the cloud environment and the ability to capture it is truly unique.
More information is available at: