Old school disaster recovery techniques pale in comparison to cloud computing when it comes to providing business continuity, according to RedBalloon general manager of development and technology Paul Keen.
In late 2011, his company, which sells experiences ranging from skydiving to pottery classes in Australia, moved its IT infrastructure into the cloud provided by Amazon Web Service (AWS), growing its server numbers from 17 to 150. As an e-commerce shop, business continuity is crucial, especially when every sale depends on it.
For that reason, Keen is passionate about disaster recovery, but not in the traditional sense. He believes disaster recovery should not only facilitate business continuity, but employee engagement as well.
"My previous company was Westfield, and every night, it took a backup of all the systems that would then be sent to another facility in Sydney, which it had provisioned for 60 seats. So out of the 600 people that worked in the company, 60 people can carry on working if the building, say, caught on fire," Keen said. "What a waste of money — it would never work. For a start, that's not business continuity.
"If you think 60 people can do the job of 600 people, then 540 of them should be made redundant."
Adopting cloud computing has not only given RedBalloon access to IT resources previously reserved for larger companies, but it also allows employees to always connect with the company, even when things go wrong, he said.
"Moving everything into the cloud is a great disaster recovery plan for business continuity," Keen said. "If one day you can't access your building, just send all the staff home, and the only thing your employees can't do is have a water cooler conversation."
Having everything in the cloud also means flexible working for employees, according to Keen.
"Whether you want to work from home, in the office, or on the train — you become a first class citizen in the IT environment, and that has huge worker engagement capabilities.
"We can give people a work-life balance — not everybody has to work in the office all the time, unless you're Yahoo."
That doesn't mean RedBalloon thinks cloud is fail-proof. AWS's Christmas Eve outage late last year has taught Keen that you can't always count on IT to work, even if it's managed by a big company like Amazon.
"I always thought if I was in two datacentres (with AWS), the chances of a region going down is so unlikely I'm not going to plan around it," he said. "I was wrong — it only took one developer at Amazon, who put in a bit of code into the load balancer in North Virginia, to bring down the whole infrastructure.
"That was terrifying for me — I could have lost a million dollars because AWS was down for 17 hours."
RedBalloon now tries to not be reliant on anything. It load-balances everything in the back end across AWS's two data centres, and has a duplication of it all. It even has two merchant bank accounts in case one has an outage.