Sergey Nivens - Fotolia
StorageCraft president talks of modernizing DR
StorageCraft President Douglas Brockett sees a future in which disaster recovery software is used to help customers flexibly migrate data to where it's most cost-efficient.
Disaster recovery doesn't have to be just about recovering from disaster.
Douglas Brockett, president of object storage and backup vendor StorageCraft, said the same software that's used to ensure a business can continue to operate if its data center is taken out by a flood can also be used to migrate scores of data to the cloud for large computing jobs. Once the need for the extra compute power is over, it can then be migrated back on premises again.
The StorageCraft president said this is already being done in the film industry, and more and more customers who experience periodic bursts in production needs will be looking for this level of cloud mobility. StorageCraft is planning to ride this wave of using disaster recovery (DR) software to achieve business outcomes instead of being just a cost center.
We spoke with Brockett about not just the future of backup and disaster recovery, but also how StorageCraft seeks to differentiate itself from other object storage vendors. With ransomware and data stores continuing to grow unchecked, Brockett said many problems remain for the storage and backup industry to solve.
What are the challenges data protection customers face today?
Douglas Brockett: All data is not created equal. If a meteor were to strike your data center tomorrow, there's a couple of things that you'd really miss in the first 15 to 20 minutes, a couple of things you'd miss tomorrow, some stuff you'll miss the next payroll cycle, but not until then, and then stuff you would never miss. And we as humans are really bad at understanding, segmenting [and] operationalizing all of that.
We allow customers to set SLAs [service-level agreements] around different types of data in their organization, and I think that's a great thing. But I think they are in need of things that can help them figure out what they actually have to protect.
So many customers today have become accidental administrators of a petabyte of data. Not so long ago, that was the sort of thing that only American Express or United Airlines worried about. Nowadays, I run into local police departments that have a petabyte of data because they have body cam footage. I run into healthcare companies and hospitals that are choking on petabytes of data. I run into schools that are dealing with petabytes of data. These guys don't have Ph.D.s in this stuff, nor the time or energy to do it. They don't have the ability to sort through all that information.
We need to take this current trend of AI and ML [machine learning] and analytics to help people find the stuff they should be caring about. Why can't I point a piece of software to my data store and say, 'Find everything for me here that looks like it's a HIPAA [Health Insurance Portability and Accountability Act] or PII [personally identifiable information] problem and apply an SLA to that?' Or, go find a file that's known to be infected with ransomware and delete it from our archives. We can't do that today.
I think those are problems we [have] to solve as an industry. They're things we're making inroads on in our labs right now.
What will StorageCraft deliver over the next 12 months?
Brockett: We have a number of things that are happening. We made a big transition from our traditional ShadowProtect product into our new ShadowXafe technology, which has gone from being agent-based to agentless or agent-based. We have new innovations around providing MSPs [managed service providers] with more tiered access capability and more remote management functionality.
In the third quarter, we've got a whole new class of product we'll be introducing. It's a product we've never done before, but I can't talk about it now.
And then, in the fourth quarter, we'll expand our cloud recovery capabilities. We see an evolution happening there. People are using backup software primarily to get stuff into the cloud for backup purposes. But that's pretty quickly shifting into people using the same technology, the same backup software we provide, as a way to manage a much more dynamic, active hybrid workload.
They have a mix of on-prem and off-prem. And during different types of use cases, they may want to move more data into or out of the cloud, and they're using our software technology to make that happen. So, fourth quarter, we'll be announcing some stuff that goes a little farther in that direction.
In what kind of situation would this kind of data movement be needed?
Brockett: The people making movies. When rendering time hits for the final production, they want that all done in the cloud, because they want it done as fast as possible, and it's a big suck of CPU resources. But when they want to do stuff in preproduction or when they're doing site selection, all that stuff happens locally, because the workloads aren't as big.
Douglas BrockettPresident of StorageCraft
So, they think about how to migrate stuff in and out of cloud as their workloads go up and down in these cycles. And as it turns out, the same stuff you do for disaster recovery ends up being a really cool way to migrate workloads in a hybrid environment.
The interesting thing about the rendering world is that it's relatively thin data going in and really fat data coming out. That's not the only world in which that happens. There are a lot of guys that are doing big data analytics, and when they've got the really big jobs, they want to push that stuff up into the cloud. Large, steady-state workloads are not economically efficient in the cloud, but the bursty workloads really are.
Amazon, Google and Azure -- they kill you for running stuff there for a long period of time. And I think people are realizing that. So, their question becomes, 'How do I get all my data up and down, make sure it's consistent, make sure I can do the same thing both on-prem and off-prem?' And the answer is your disaster recovery software was designed to do just that. The problem is it's not sold, marketed, positioned and doesn't have the UI that makes you think about it that way.
Do you see this becoming a trend over the next two years?
Brockett: I think the people who are most savvy about this are cloud guys who have this view that if they're spinning stuff up on the cloud for DR purposes, maybe it should just stay there. They're at the point where they're trying to make their cloud as sticky as possible so that anything that ever touches it ends up migrating and staying there.
I think you have a generation of CIOs who were early adopters of cloud that are seeing really big bills for doing stuff steady-state. So, the winning strategy for these CIOs is getting as mobile and flexible as possible between on-prem and off-prem.
How do you compete against other object store vendors, like IBM Cloud Object Storage, Cloudian and Dell EMC?
Brockett: I think the key thing for us is that we don't make people even need to understand that they're dealing with object storage. You buy a $10,000 or $15,000 box, you flip on the switch, and in five minutes, you have an SMB and NFS share.
We sneak in there without you having to do anything for any app. We look like you just bought an inexpensive NetApp or Data Domain, while giving you all the scale-out benefits of object store. And that doesn't require any developer getting involved.
Channel guys are not typically writing a lot of code or converting a lot of apps from one format to another. They're largely selling other third-party apps. They don't have to go get Ph.D.s in the product before they deploy it. That's how we win.
StorageCraft acquired Exablox object storage in 2017, and it's now a core part of its business. Do you see more acquisitions coming?
Brockett: We don't have anything to announce right now, but we are always looking for stuff to add to the portfolio that brings value to end customers and channel partners.
We live in a very dynamic environment in which new stuff is coming out every day. And to imagine that you and your company, organically, are going to be smart enough to figure it all out, that's a hubris level that I don't have. Nobody has a crystal ball.
I'm pretty happy with where our product lines are right now. Our pattern tends to be to buy companies that are looking over the horizon a bit. We want guys that have things that will be a big market three to five years from now. We'll use acquisitions to jump-start ourselves down that path.