Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

The promises of IoT and data management strategy

Zafar Anjum | April 24, 2015
In this wide-ranging interview, Mark Bentkower, Director of Systems Engineering, ASEAN, CommVault, talks about IoT and its promises in the context of other technologies and imperatives such as cloud, big data, virtualisation, security and data classification

We're trying to give customers the opportunity to do a lot more than just backing up. We want to let them use the data intelligently in a number of different ways. Disaster-recovery, edge and virtualization are the 3 things that my customers in this region want to talk about the most. 

What is edge?

Edge is any devices that is on the edge of the data center - laptops, mobile devices - multiple devices that people carry.

Everybody is trying to virtualize as much as they can because of big cost savings. Virtual machines, which are very easy to make, bring a lot of difficulty because they take up CPU and disks that aren't being used anymore. So we're doing virtual machine life-cycle management. CommVault identifies inactive virtual machines and asks customers if they are keen on archiving or getting rid of the machine. Enterprises would also want to synchronize virtual machines and put them in the cloud as well. They want to make sure when one machine goes down, the other picks up. 

The other thing is that enterprises tell us they have a 10-year-old Windows 2000 machine that is still running something business-critical but they can't buy the hardware to replace it or upgrade it. Other than simply backing up for you, CommVault can run something called Virtualize Me, which allows us to turn a physical machine - Windows or Unix - into a virtual machine. Now, you don't have to worry about the hardware anymore because you got a virtual version and you don't worry about the software anymore because it's working. We're giving companies opportunities to take some of these problematic, old legacy machines and turn them into a more modern platform. 

So your business is basically sitting between real physical infrastructure and the cloud. You're basically providing the connection, right? 

We look at ourselves as giant virtual data store. We look at local disks, SANS, NASAS, tapes and clouds - both public and private clouds. To us, they are magnetic libraries, it's all obscured. We're just moving from one library to another library and as long as the user is understanding that there are physical limitations to how fast disk spins and networks go as long as you're not trying to do something faster than it can be done, it's not a problem. With data classification, we help them understand what the data sets are that are most important and that they need to invest in a really fast hardware. But to not waste it and buy hundreds of terabytes of super-fast this disk when there's really 15 terabytes that's mission-critical. 

Once you virtualize the data, it is going to be there forever or will there be a possibility of loss? 

A virtual is just a giant disc, with an engine sitting under it. There's always the possibility of losing data. By keeping multiple generations of that data in multiple places, you will be able to get around the problem of losing data. By keeping the data on high-quality discs you have the safety of the hardware checking, making sure that you're not going out to have corruption or data loss. By keeping multiple generations in archive, you save yourself from somebody accidentally going in and erasing all of that data. 

 

Previous Page  1  2  3  4  5  6  Next Page 

Sign up for CIO Asia eNewsletters.