Is it time to think about disaster recovery as a service?

This article was originally published on Technology.Info.
As part of our continuing strategy for growth, ITProPortal has joined forces with Technology.Info to help us bring you the very best coverage we possibly can.

Continuing the series on

Building the Hybrid Cloud

, we speak to IP EXPO Europe 2014 speaker,

Joe Baguley

, chief technology officer for EMEA at

VMware

, about why organisations are increasingly turning to hybrid cloud technology for

disaster recovery

.

In mid-April, virtualisation specialist VMware announced VMware vCloud Hybrid Service - Disaster Recovery, a service that provides customers with a continuously available recovery site if their own, on-premise VMware-based environments run into trouble.

The new disaster recovery service offers a recovery point objective (RPO) of 15 minutes, at prices starting at $835 per month. The aim, according to

VMware

executives, is to provide customers with “a simple, automated process for replicating and recovering critical applications and data in a warm standby environment at a fraction of the cost of duplicating infrastructure or maintaining an active tertiary data centre.”

For some customers, it may also represent an opportunity to protect applications that have previously been omitted from Disaster Recovery plans for reasons of cost and/or complexity. As we heard in

Article 1 of this chapter

, disaster recovery is fast emerging as an important

hybrid-cloud

use case.

Technology.info sat down with Joe Baguley, chief technology officer for EMEA at VMware, to discuss the new service and to ask him: is the way that companies look at

hybrid cloud deployment

starting to mature?

Q: So, Joe, what’s the thinking behind VMware vCloud Hybrid Service - Disaster Recovery?

A: Our thinking here is simple: hybrid cloud should be seamless. It shouldn’t be a migration - just a seamless click within your existing environment. The whole point with this service is to provide an an easy and simple way to back-up an existing VMware environment to the cloud, so that in the event of a disaster hitting their own data centre, a customer can quickly can spin up those environments in the cloud.

Q: And where exactly will those systems run - in VMware’s own data centres?

A: Sort of. What we do as a company is rent data centre space from the likes of Equinix and Savvis, but we use that space to provide a managed service to VMware customers that is entirely operated by us. Plus, there’ll also be a wide range of VMware partners offering the service to customers.

Learn more about building the Hybrid Cloud at IP EXPO Europe 2014

Q: So what does the introduction of this new disaster recovery service tell us about wider market demand for hybrid cloud services?

A: I think it’s fair to say that, initially, hybrid cloud was largely seen as a good place for dev/test environments [development and testing environments where software developers would build and try out new applications before deploying them in-house]. But many of our customers have quickly identified the hybrid cloud’s potential for disaster recovery (DR), too. We’ve actually found some interesting cases where customers come and stand up DR in our hybrid cloud, do a failover test and then realise that the system runs better in our hybrid cloud than it does on their own premises! A few have even switched to using our cloud as their primary site and their own for Disaster Recovery.
Others have said they were interested in the idea of ‘cloud-bursting’, so that on-premise apps can take advantage of our capacity during periods of peak demand - but many have quickly found that an awful lot of recoding of applications was needed in order to enable them to do that. But where we are now with hybrid cloud is that, because we’re standing up exactly the same technology stack in our data centres as we sell to customers to run in theirs, it’s a relatively trivial thing to migrate workloads between the two - they’re effectively the same environment. As a company, VMware has around 45 million VMs [virtual machines] running in customer sites today - and we’re giving those customers a place where they can ‘drag and drop’ those VMs if they need to do so. As we discussed in this chapters article on Hybrid Cloud Considerations, cloud bursting is another emerging hybrid-cloud use case.

Q: So where does that leave the whole issue of cloud interoperability and the industry standards effort that’s going on around hybrid cloud computing?

A: Well, it’s an interesting question. My answer would be that there are still many cloud interoperability issues to tackle and several major standards efforts underway - but when I sit down day to day with our customers, a lot of them say to me: “We’d love to pick an industry standard and go with it - but we can’t choose one yet, it’s just far too early.”
The one standard they do know - or at least, a de facto standard - is that they already run VMware today, so for now, that’s the technology they’ll stick with. They don’t want to pick an industry standard that’s not going to win the race in the long term.
But that’s not to say that VMware isn’t part of the wider IT industry standards effort. We’re proud to say we have a very high commit-rate to OpenStack and we are 100% backing OpenStack as we go forward. Customers will see us develop more and more interoperabilities as that platform develops but, at the moment, it’s still early days. That’s just the way it is in this industry: technology development moves faster than any standards body can.

Q: But isn’t that luring customers into putting all their eggs in one basket - a VMware branded basket?

A: It’s also about having ‘one throat to choke’: customers can buy their licenses for on-premise and their credits for hybrid cloud in a single transaction, get the same support line and avoid the need to make changes in personnel and skills. We’re working to make hybrid cloud a natural extension of all the stuff they already do.

Topics