I’m back talking about Actifio, after my previous visit in their offices, and this time in San Jose joining the other delegates at TFD15. In a previous post I was curious about what I was going to find this time, comparing with what I learnt last time. Expectations were better than I imagined.
The main concept in Actifio is that “Data is a strategic asset”
Something I had in my mind was presented by Ash Ashutosh in a clear way, at least for me: some years ago, business was driven by IT: IT moved at its own speed useful for cost saving, and the business followed it. Today, the process is inverted: Business drive IT, cost is nomore the driver but speed&agility are. Speed is set by business, and an army of inside developers could accomplish these needs, deploying on cloud. This last part can be summarized by an overused word: DevOps, together with analytics.
The difference between the 2 models is represented by the following image:
Analytics of the developed applications gain a huge value for the company, same for DevOps. Ash makes a clear example: Friday night, system is down, in the first case all you can do is restore few applications, either with DR or, in the worst case, restore archived data; in the second case, I have plenty of developers available, my concern is on the business application, not on the system ones: they can check and manage the business applications via API, they aren’t relying on your on-premises infrastructure.
This is the new paradigm, following the previous example: cloud is no more simply the place where applications reside, a datacentre, but a platform to deliver applications.
This paradigm consists on three important elements, everyone has its place in the cloud (well… “should”):
- Infrastructure, moving to IaaS – easy to move
- Applications, moving to SaaS – easy to move
- Data: actually, it misses the corresponding service in the cloud, and it’s the real core of the whole stack, it’s the “lifeblood of the business”
Solution is having a so called “Enterprise Data as a Service (DaaS) that makes the data stateful as before, but easy to move instantly all around the cloud, that is all around the world, and that connect the previous 2 elements.
All the related functions to be operated on our dataset stack can be split all around several and different cloud providers, choosing the one that, in our opinion, is better on: one for DR, one for backup, another for Windows applications and so on.
This shows a kind of serverless infrastructure, but also a storageless one
Of course, this whole discussion is applicable to hybrid cloud. We all agree that the companies switching completely on a public cloud are just a few percentage, for several reasons out of the scope of this discussion.
That’s the new stuff that I found after my last visit at Actifio. Beside to tools that improve resilency and availability, beside facilitating building of HQ applications faster, the new feature is enabling an enterprise hybrid cloud.
Sky Platform 8.0 was built around this new concept. It’s cloud native (it couldn’t be different) currently supporting AWS, IBM Blue Mix, Azure, Google and Oracle. It allows cloud mobility (no lock-in). It uses cloud object storage, and this is revolutionary. Think at an Oracle DB using object storage for its tables. It can use a cloud catalog to support scaling using metadata, via API. Lastly, if I, user, add a great value to the data loaded in the cloud, you can expose this value through cloud engagement.
I consider this new paradigm about data as a game changer. I’m sure we’ll be able to discriminate in the next future who will adopt a DevOps model like this one, and who will use a traditional – right now we can also do the same wit who use DevOps and who doesn’t, actually.
I want to go deeper technically in my following post, so keep an eye on my blog. I’m sure you’ll be amazed at least as I was.