Last July, Dimension Data's Australian cloud went down for over 24 hours. Now the company says its assessment of the incident found those who suffered the most had themselves to blame, to a degree.
Speaking today at the launch of the company's new government cloud, cloud general manager David Hanrahan said those impacted by the outage fell into two categories.
Those who felt most pain, he said, “had not architected for availability” by replicating data and applications to either their own premises or to other clouds.
Customers who had “taken an enterprise architecture approach approach and mapped their applications from top to bottom and planned accordingly” experienced less pain as a result of the outage.
Hanrahan said the customers who experienced most pain assumed that when workloads go into the cloud "magic happens". Since the incident, he said, DiData is having many more conversations about proper backup and resilience regimes. He added that the crash, thought to be a result of an EMC failure, “highlights that delivering these highly complex as-a-services depends on interoperable technologies.”
Hanrahan's mostly right: it is possible to architect an application so that when a supplier fails a cloud provider redundant systems will kick in to protect customers applications and data. But the need to do so is not often mentioned in the rush to point out cloud's low price, elasticity and speed of deployment.
This incident, and other like the recent Azure outage, will surely focus would-be and current cloud users' attention on the real requirements – and costs – of adopting cloud. ®