Cloud computing was front and centre at the British Columbia Technology Industry Association’s Techforum luncheon series, held at the Sutton Place otel in downtown Vancouver. David Mitchell Smith, a Gartner Fellow and VP from Gartner Research, addressed a packed audience who wanted to know what impact cloud computing will have on the IT sector, what service markets will emerge and how companies can leverage the cloud to support their IT and business strategies.
Though cloud computing can be viewed from multiple perspectives, Smith defined cloud computing as a “style of computing” which enables flexible and elastic capabilities provided as a service using existing Internet technlogies.
But while users see the cloud as an amorphous thing that you can draw from, much like the Web, from the provider point of view the cloud boils down to actual hardware and software. IT has to understand both those perspectives, he said, because while historically IT has been a provider, the cloud has morphed its role. IT has to be able to juggle, providing what it can and then sourcing other tasks to the cloud when necessary.
There are certain traits, Smith said, that define a cloud service, and the first is in fact service. There’s a level of abstraction between the provider and the consumer that delivers resources. The cloud is scalable, elastic, and shared, and is metered by use.
Interestingly, Smith explained that not everything actually needs to be in the cloud. A hybrid model “will and should dominate for the next 10 years,” he said. There’s also a tendency to define anything remote as cloud computing, when the cloud is fundamentally a service model.
Smith invoked the Gartner “hype cycle", the inevitable peaks and valleys every technology goes through before it becomes generally accepted. Cloud computing is currently at the peak of the hype cycle, with many technologies adopting the model whether its the right fit or not. But eventually, he said, the technology will mature and spread out where it’s needed.
Smith pointed to Amazon’s Web Services as a good example of an all-encompassing cloud service that gives IT an entire system to leverage. Google, on the other hand, has reams of infrastructure but does not give the user access to the system at the same level as Amazon. Google offers more Software as a Service (SaaS) modeled apps than Amazon, and in doing so adds in a layer of abstraction between consumer and provider.
Microsoft is, as usual, pretty complicated. Microsoft offers ways to let people run off a service, but also through mobile, through the Windows OS and elsewhere. Smith said Microsoft’s strategy is by far the most ambitious and has many dimensions going beyond the “as a Service” model. But it remains to be seen if this complicated arrangement, though technologically impressive, will be enough to distinguish Microsoft from its cloud competitors.
So why should people consider cloud computing? It comes down to cost and capability. A startup costs much, much less to build now because of Amazon’s cloud infrastructure than ten years ago, when a company had to build everything themselves. But companies are also using the cloud for proofing their ideas, serving applications, email, collaboration, development, and SaaS.
But there are risks in the cloud, such as offloading your data to a service provider, and then worrying about availability of the service, security and integration with current systems.
Future uses, Smith said, include cloud service brokerages, which find the resources needed by consumers and enterprise and create a custom solution using available cloud services. This opens up new capabilities, such as aggregation, community management, business social networking and arbitrage.
Smith said the four things a company can do with cloud computing today are to compare their capital expenses to cloud providers, find workloads they can experiment with by moving them to the cloud, possibly even wrapping existing applications into the cloud. Finally, companies can begin to adopt cloud email and collaboration in order to take small steps into the cloud while still measuring whether the risks are worth the possible reward.