By 2030, the amount of data is forecast to have grown to 1 Yottabyte (a lotta bytes, for sure...think 1 followed by 24 zeroes). That was the prediction of Martin Saddler from HPLabs, who I heard make a presentatation in London a few weeks ago.
Chris, excellent question. Actually, enterprises having build an "enterprise bus", in other words, an integration architecture amongst their companies, using tools such as Tibco or JBOSS to name a few, have a much easier task as they can extend that architecture to their cloud based environments and only need to develop/adapt one interface. The costs of information transfer will be there anyway, but you can get a reasonable assumption of the volumes. Companies not having such approach should first implement one of those to facilitate the integration. Now, is that cost part of the cloud journey, or is it an approach to facilitate maintenance of the existing environment and reduction of existing integration costs? Good question.