Is the Cloud Being Over-Sold?


So exactly how much hype surrounds cloud computing these days? Actually, no more than usually accompanies a new technology in the earliest stages of introduction. Think back to virtualization, SOA, Web 2.0-heck, the Internet-and it's clear that the initial round of hype is part of the normal pattern of IT evolution.


But that's not to say there aren't a lot of overly bold claims in the cloud or that the technology isn't being oversold as the true answer to enterprise nirvana. Quite the contrary -- there are a number of challenges to be faced when either deploying your own internal cloud or tapping into a third-party provider.


Jian Zhen over at LogLogic spells out a number of chief concerns in this blog. His five key challenges include data governance and information lifecycle management, management of cloud resources themselves, performance and availability monitoring, reliability and available, and, of course, security. None of these issues are insurmountable, mind you. It's just that in this early stage of the game, these are the weak spots that must be addressed if the cloud is to become the dominant architecture that supporters expect.


Another major drawback is one of simple perception, according to Jeffrey J. Hardy at management firm SmarterTools. The typical cloud is presented as a loose collection of applications and resources that can be cobbled together in amounts and configurations to suit specific needs. The fact is that all cloud resources are located on physical machines somewhere, meaning the cloud is both finite and prone to all the ailments that the real world has to offer.


Once deployed, there is a strong potential for conflict between IT and business units as to who should use the cloud, and under what circumstances, according to rPath's Billy Marshall. As IT seeks to hold down virtual sprawl by limiting the number of VMs in the enterprise, business units will simply turn to the cloud to get what they need, costs be damned. Automated application management and placing IT in charge of cloud deployments will help, but it will take a coordinated effort to keep everyone in line.


In the end, however, the cloud is coming whether you want it or not, says Simon Wardley of UK software firm Canonical. Once the ball gets rolling, enterprises of all stripes will take to the cloud if only to avoid being left behind as technology marches on. Psychologists know this as "The Red Queen Effect," and it will be virtually impossible to resist once the tools, both hardware and software, for building and deploying clouds become componentized and commoditized.


On this I half-agree. I do think the cloud in some form is inevitable, but anyone who claims to know how it will evolve and what functionality it will have in the future is grasping at straws. Yes, it's coming, but how it will all shake out is anybody's guess at this point.