CenturyLink: Why complexity, not the cloud vendor, locks you in

It continues to be an extremely busy time for CenturyLink Cloud. Amidst a plethora of news and acquisitions, one stood out: the buyout of disaster recovery as a service software provider DataGardens.

As this publication has previously examined, cloud disaster recovery is certainly one of the more popular buzzwords in the IT industry right now, yet a lack of clarity still pervades. There are degrees of severity, from a short outage to a full blown DDoS attack, and there are different strategies vendors take; Verizon’s decision to implement a planned outage in January was roundly panned by the industry.

David Shacochis is VP cloud platform at CenturyLink. Speaking to CloudTech at Cloud Expo Europe, Shacochis describes disaster recovery as a service (DRaaS) as “a good headline”, but argues it misses the point.

“It’s certainly an important conversation starter, but really what it’s all about is workload portability,” he explains. “If you have workload portability and flexibility, and the ability to move workloads around and keep them – if you have that flexibility then that’s a risk mitigation, and that risk mitigation is ultimately what disaster recovery and continuity of operations is all about.

Shacochis continues: “Disaster recovery, more and more in the cloud age, is about architecture and design, to mitigate against the need for disaster recovery. DR as a service is a great way to start a conversation, [but] I think we’re increasingly getting to the point where disaster recovery declarations are not really what the industry needs to hear.”

The proof is in the pudding, Shacochis argues, in the new wave of application development and architectures; the applications redesigned to be resilient in the age of cloud are what people need to hear about instead of hard luck stories.

“A lot of the state management, session management and consistency architectures that modern application architects are designing for are starting to remediate the need for a lot of that,” he says. “Every graduating class of computer scientists starts to buy in more and more to that architecture, that certain way of designing and way of thinking.”

As Marc Andreessen once wrote, software is eating the world. Nowhere is that more appropriate than in cloud disaster recovery. Whereas once the DR strategy was storing a pile of kit in a data centre somewhere, software is disrupting it completely. “It’s very easy to take a copy of a cloud application and copy it to another cloud provider and keep it there as your hot standby,” Shacochis notes.

Naturally, prevention is better than cure; it’s important to have disaster recovery implementation in place, but it’s better if you don’t have to resort to it. Hence the importance of the cloud exit strategy.

For CenturyLink, whose acquisition of DataGardens was a mix of hiring the talent involved as well as the product, building services that are easy to migrate in and out of is “fundamental” to what they design. The oft-reported concern, of cloud vendor lock-in, is a misnomer according to Shacochis. It’s not vendors that provide the lock-in – it’s complexity.

“You can not be in a cloud,” he says. “I’ve seen colocation cages and environments that just make your heart cry. It’s just a tangled mess. They would simply have to rebuild it somewhere else in order to ever let go of that environment.

“We particularly think that cloud computing is fundamentally a lock-in free environment if done right,” he continues. “I think there are some cloud environments and some cloud platforms that are getting so clever and innovative, like what Amazon’s doing with Lambda or some of their proprietary modular services.”