Cloud computing is extremely powerful. Many companies dedicate a large portion of hardware processing power and data centre storage for this type of application – however, for many developers it is not easily accessible or usable as a viable means of centralised storage or application sharing. This is why there needs to be a standardised platform available for developers to easily access their cloud’s that may be spread over multiple data centres or clouds. The University of California (Berkeley) is tackling this problem with a great deal of interest. They are attempting to modify existing database programming languages and integrating the modifications into a single client platform that is able to access a variety of clouds.
Database languages aren’t really a difficult application domain – however, the diversity of languages out there make it difficult for a novice developer to really get into the specifics of a number of different languages and many novice or smaller developers stick to a single database platform because it is not feasible for them to employ database consultants that specialise in any single database language. If you consider the range of clouds out there that are relevant to social networks, communications providers, search providers and any other data storage relevant niche then you can easily see the range of specific technologies required to access clouds from such a variety of sources.
You also have to keep in mind that cloud data is not exactly static. It is dynamic because it changes constantly. Having a single software tool (such as that developed by Berkeley) will enable developers to access a variety of clouds and scour for changes in the data in a quick and efficient manner. It will also allow them to develop a range of applications that are usable on a variety of domains rather than just a single cloud.