Strategizing to preserve business continuity and prevent downtime is a continuing quest. From tapes to floppies to flash, companies have stored data in one form or another to restore in the event of catastrophe. Now, as remote operations become the norm, what is considered mission-critical has changed. So too has the speed of recovery.

In a world where the only acceptable downtime figure is zero, solutions must evolve to keep up.

“What we’ve seen is this genre of the ‘lights out data center’ … has become absolutely critical to operating a business today,” said Doc D’Errico (pictured, left), vice president, office of the chief technical officer, at Infinidat Ltd. “This whole concept of business continuity now encompasses not only the operations equipment … but it also encompasses applications that people need access to that they may not have thought of as mission-critical before.”

D’Errico and Ken Steinhardt (pictured, right), field chief technical officer at Infinidat, spoke with Dave Vellante, host of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, for a digital CUBE Conversation on strategies and solutions for maintaining business continuity under the constricts of remote operations. (* Disclosure below.)

[Editor’s note: The following content has been condensed for clarity.]

You are both long-time storage industry experts. What is the state of business continuity from your perspective?

Steinhardt: Historically, organizations have focused far too much just on traditional disaster recovery with things that have been around a long while, like backup, and all too often haven’t focused toward the technologies that really do keep the business running without human intervention if something would ever go wrong.

I was on a chief information officer roundtable this morning, and there were a couple of interesting comments that really stuck out to me. One executive was commenting on how much people are working from home now. He said they’d recently hired about 250 people, and all of them are going to be remote workers. One of the other CIO’s said: “Remote work requires always-on infrastructure in this day and age, and it’s just a whole new way of having to make sure that businesses are operational and workers can do what they’re supposed to do.”

D’Errico: Companies are accelerating a journey. They’re also looking now at this concept of multiple active sites. When the financial industry was struck [by the impacts of COVID-19], they were looking at some very significant changes in their operational paradigms because they realized that the system going down is only a small percentage of the problem. The people impact is far worse than the operational procedures.

What they would do is typically build out multiple sites and rotate the applications between them; now they’ve started this journey of having the applications running simultaneously in multiple sites accessing the same data sets. It’s not a brand new concept, but the technologies have improved significantly over the course of the past decade, and the introduction of our InfiniBox Active-Active solution a couple years ago even brought it to an entirely new level.

Can you give us sort of a “data protection 101” review in the context of this discussion?

D’Errico: The important thing to look at when you think about the different types of technologies that you can apply to solutions is that some of them apply to specific equipment failures and some of them apply to data failure. I separate equipment from data in the sense that data could be corrupted in some shape or form. It can be through malicious attack — ransomware as an example — or it can be incidental, you know somebody pressing a wrong button. It can be an operational procedure, perhaps another system failure that causes a change in the data or corruption in the data that makes it essentially unusable.

So, we have to start with what is the recovery point objective — the RPO. If you think that time zero right now is where the failure occurs, you work backwards. How far back can I go and still sustain my business? Each of these technologies that we’re talking about gives you a different RPO — it’s like rewinding a tape back to a point in time.

People don’t want to lose data. But when you ask a customer how much data are you willing to lose, they say none. So you say, well how much you willing to pay? There’s a tension in the dynamic. How do businesses measure the impact of disaster recovery?

Steinhardt: It’s the traditional trade-off between RPO, [recovery time objective] and cost. With RPO the objective would be to get as close to zero data loss as you could possibly get. RTO measures the time associated with how long will it take you to get back to your acceptable level of RPO.

Historically, the closer you get as you approach zero RPO and zero RTO, usually the greater the cost goes up. It’s always been the eternal trade-off. It’s sort of like if you want to buy a car. If it’s good and it’s fast, it won’t be cheap; if it’s good and it’s cheap, it won’t be fast; and if it’s fast and it’s cheap, it won’t be good. There are many aspects that you have to consider in terms of what is the service level that the business requires, and do we have solutions in place that can actually give us what is the real service level the business requires if something were to go bad.

Image: Infinidat Ltd.

D’Errico: A business may have different service levels for different applications right, and you never really know what that is. For example, I was working with a university a few years back, and you’d normally think they’re worried about their grading systems or about their financial systems. This particular university was worried about their golf course reservation system! It was their number one mission-critical application, because it directly impacted their alumni and had a direct correlation to the incoming donations.

When you think about the different types of replication technologies that are available, you look at what multi-site is really doing. Multi-site is giving you some level of synchronous replication so that you have an RPO of zero; it still may not be an RTO of zero, but it will be darn close to it. But more importantly it’s giving you an additional site to really maintain that RPO of zero in case the disaster radius is even further away. Now this isn’t going to prevent any type of malicious intent; it’s not going to prevent the ransomware case and things like that. But it’ll certainly prevent the catastrophic failure of the data center.

Active-Active now gives you the read-write capability, and our multi-site implementation gives you the ability to have a simultaneously running instance of an application in multiple data centers reading and writing from the same data set. What that gives you is not only an RPO of zero, but an RTO of zero.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s CUBE Conversations. (* Disclosure: Infinidat Ltd. sponsored this segment of theCUBE. Neither Infinidat nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE