By SunGard Availability Services Cloud Consulting
A reminder from the NetFlix Christmas Eve outage…
It seems to be human nature to look for “silver bullets”. With each new advance, technologists hold out hope that the challenges of the past can be put behind us. “This time it will be different!” But more often than not each advance brings with it a new set of challenges in exchange for ones it solves. This is certainly true of cloud. It can reduce your risk, or increase your risk – all depending on how you approach it.
We must always anticipate service interruptions no matter what the platform or provider. They are going to occur, just as the sun is going to keep rising. Whether simple human error, code bugs, the next hundred-year storm or an errant backhoe up the street ripping out your fiber, something is going to happen, and often at the worst possible time. For many organizations, cloud services will provide immediate gains in agility and scalability. On the other hand, very few organizations are going to find instant adequate fulfillment of their availability or resiliency needs by simply placing resources in a private cloud or with a public cloud provider.
The good news is that when done right, cloud solutions absolutely have a place in meeting or beating availability needs along with providing desired agility and economic improvements. BIA’s, risk assessments, enterprise architecture and the other disciplines that have served us well for all these years in producing quality technology estates are as relevant to cloud as they have been to all of the platform and operating models that came before. The leading cloud providers and solutions all offer a broad set of enabling services and capabilities to facilitate meeting the availability requirements that those disciplines help us enumerate.
The more mature platforms even offer a reasonably comprehensive set of features and options to build platforms that can run in multiple places as needed to ensure the ability to work around a problem in one of them. Zones within regions within geographies, underlying snapshot and replication of storage resources, workload management tools for mobility within and across cloud providers, and experts who specialize in putting it all together; these capabilities and more are on the table.
There is always an uproar when a name provider experiences an outage of any kind. We often hold them to a higher standard after all. But as unfortunate as their outages are, the end-user service disruption is really the responsibility of the cloud consumer. It is up to each individual client to work with their organization, its partners, and their solution providers to put it all together creating and ensuring a comprehensive approach to meeting the full scope of relevant business requirements. Equally important is that we are learning iteratively when we encounter an unexpected error, getting better every time. Outages are a good reminder that cloud is only as good as you make it. And that is where the fun begins…