Restore as a service – Backup is good, but restore is a lot more useful. I talk to W. Curtis Preston, “Mr Backup”.
Druva is all about “data management as a service” but, for most of us, that probably means “backup/restore as a service”. There is, of course, more to it, and to Druva, than that. Druva’s business solutions cover compliance monitoring (including GDPR), ransomware recovery, operating system migration etc. – as well as, e.g., endpoint backup and cloud disaster recovery. As it says on its website: “Your data. It’s everywhere. Scattered across cloud applications, data centers and end-user devices. Pushing data protection and management to its limits. The Druva Cloud Platform provides a simple, scalable, and on-demand approach to take back control of your most critical data – all as-a-service”.
The “as-a-service” bit is important as being able to protect and restore data is becoming the subject of regulation as well as being an essential part of managing business continuity. Asking customers to re-enter their personal data is a bit of a no-no (and might be a very expensive no-no, post-GDPR and similar regs.).
Actually achieving effective and timely business continuity risk management is far from trivial, however. If you buy it as-a-service, you also buy into the specialised expertise and experience of the service provider, gained over a wide range of scenarios, including some that you haven’t met – yet.
The sort of non-trivial issues you might want to discuss with a service provider include the GDPR “right to be forgotten” – does it apply to your circumstances (this is a legal issue, not simply a matter for common-sense arguments and your opinion); and, if it does, what about the data in your backups and relational databases? You may not be able to delete individual data items without destroying entire archive and database structures, and recreating them from scratch without a particular customer might be too slow and error-prone. Possibly, a restore process that deletes “forgotten” data before anyone sees it might be sufficient, but this hasn’t been tested yet – it is something Druva is thinking about.
One new Druva feature just announced, which I rather liked, is support for the Druva Cloud Platform on Amazon Web Services (AWS) Snowball Edge. This “allows enterprises to freely move data between on-premises and cloud without the lock-in, complexity, and cost of legacy hardware”, which (amongst other things) will speed up and de-risk cloud transition initiatives.
What Snowball Edge is, essentially, is a 100TB hardware cloud cache at the cloud edge, managed by Druva and AWS and transparent to AWS customers. It supports Amazon EC2 and Lambda processing, which Druva uses to shift data onto and off the cloud, but the customer just sees a cloud that can operate with something like on-premises performance even for very large amounts of data (100 TB is mentioned) – and, of course, with Druva encryption services (even Druva can’t read your data, but it can advise on key management strategies without a single point of failure) and global de-duplication (only one copy of any piece of data is ever stored).
Cloud, to me, is as much a cultural thing as a technology thing – and thinking in terms of some form of data-management-as-a-service (as opposed to data management technology in-house) is a no-brainer, in this culture. Of course, there are issues if you lose control of your regulated data; but Druva seemed to have anticipated all of my concerns in such areas. The only issue I came up with was scaling small – Druva only starts with 10-user companies and scales up a very long way. Not selling to one-person companies (so, no trial single-user editions available on Web download), is a marketing decision, not a limitation in the technology, according to Curtis Preston of Druva, who was telling me about his new home there. I can see why Curtis joined Druva. I like its cloud story too.