One of the problems with cloud computing is it is so flexible. We’re still in the early stages of the technology and – pardon the pun – the sky’s the limit on what can be done.
Still, some people think there are essentially two clouds: One with offerings for enterprise workloads (with infrastructure resilience assured), and another for fully-native cloud apps (where resilience is NOT assured because there are no high availability services).
What’s the difference? Well, the discussion was sparked by a blog by Massimo on IT 2.0 following an outage in April on an entire Amazon AWS region. In his blog Massimo argues that AWS essentially offers and efficient computing platform that doesn’t have HA services. If you deploy your app in one of its availability zones, there’s no promise it will fail over into another zone.
But in a blog last week Gartner analyst Lydia Leong offered an opposing view: Customers won’t have to choose between one or the other. Cloud market leaders will offer a range of infrastructure resources – some will be more resilient, and therefore more expensive, than others – she argues.
Remember, she says, enterprises have huge numbers of legacy apps they will need to host. But also, she writes, enterprises still write non-native small cloud apps that don’t need to be resilient.
It’s an interesting piece and the links within it should be considered by organizations looking ahead to what the impact of cloud computing will be.
Sponsor: IBM Canada Ltd
IBM Blueworks Live, the roadmap to tackle process improvement
This IBM whitepaper examines how companies can use the Blueworks cloud-based, wiki-style collaborative platform to gain the clear understanding of business processes necessary to achieve true enterprise agility, identify process bottlenecks and gaps, and take the first steps down the road to business transformation.