Nine Cloudy Challenges for Databases
It seems that the heady days of cloud computing are coming to an end, which means the real work is about to begin.
As more organizations deploy both private and public cloud architectures for extra storage, backend software support and other functions, attention is turning toward integrating these resources into an overarching, orchestrated data environment. But while most observers were initially skeptical as to whether the enterprise was ready for the cloud, it could very well be that the question should be reversed: Is the cloud ready for the enterprise?
According to Unisphere Research, nearly 40 percent of organizations worldwide have established private clouds, and about a quarter use the public cloud. What’s interesting, however, is the speed at which existing cloud users are moving past simple storage applications into more advanced architectures like PaaS and IaaS. If this trend continues, it won’t be long before the cloud becomes the primary means of supporting critical data operations, producing changes in all aspects of the data environment from infrastructure to managerial skillsets.
Of course, this wouldn’t be happening if there wasn’t a large pot of gold at the end of the rainbow. As ActiveState Software CEO Bart Copeland points out, the benefits of moving to a full cloud-based PaaS architecture range from dynamic resource scalability to more efficient and agile development and provisioning. Companies that can master the intricacies of PaaS will become more innovative and will be able to get products and services to market much faster than their virtual brethren, which means not just lower costs but increased productivity, and profitability, as well.
But is it really that simple? Is the key to vast riches and glory simply a matter of deploying PaaS, IaaS or any of the other as-a-service architectures? Possibly, but the devil, as they say, is in the details. Freeform Dynamics CEO Dale Vile makes a good point in that cloud environments will likely be as unique as the enterprises that deploy them, and the lines between what is and is not platform or infrastructure will be very blurry. With various platforms employing different means to access resources, execute applications, pull data and perform other tasks, the challenge in building the “right” architecture will be just as great, if not more so, than it was in the physical or virtual worlds.
And in the end, there is no guarantee that any of this will actually fulfill enterprise expectations, says tech advisor Larry Carvalho. With the cloud-hype machine running at full steam over the past two years, those expectations are naturally quite high. But with architectures like PaaS designed to abstract underlying infrastructure in the first place, how is the enterprise to know they are actually getting what they pay for? Greater visibility into resource usage will be a must as reliance on the cloud increases, but it is questionable whether many providers are willing to provide such insight into their inner workings, even for valued clients.
The cloud, then, represents the classic push-pull dilemma for the enterprise. On the one hand, the front office is pushing for more productivity and lower costs while IT is pulling for clear architectural development strategies and, above all, solid security and availability.
The good news is that the cloud provides a clean slate on which to build a flexible, dynamic environment suitable for a decade’s worth of software development and data operations. The trick will be in laying a foundation that is flexible enough to support the many twists and turns that the high-level architectures will undoubtedly take in the coming years.
