Want Cloud Control? - Bubble Wrap Your Apps with a Private PaaS
Wikipedia defines crowd control as—not surprisingly—“the controlling of a crowd, to prevent the outbreak of disorder.” Picture stanchions, yellow-jacketed sports-event ushers, maybe even police with helmets and shields. (Here in Vancouver, Canada, we vividly recall the chaos that followed the Canucks’ 2011 Stanley Cup loss.) In its many forms, crowd control efforts lessen disarray, and aim to preserve order. But when it comes to cloud computing, effective “cloud control” isn’t something you can take for granted. I define cloud control as “the controlling of data and applications in the cloud to prevent the outbreak of mismanaged information, inaccessible applications, and security breaches.”
For the real-world enterprise, the cloud promises flexibility, efficiency, and convenience. But those dizzyingly seductive promises come with reality-check hangovers like data security and privacy risks, not to mention diminished peace of mind. Maintaining private on-premise bare-iron server farms is an expensive and not-particularly scalable solution for the long term. Deploying to the cloud is an attractive alternative. But handing off mission-critical apps to a third party service provider takes control of data and applications out of the hands of the most important person in this equation: you, the owner of that data and application.
And that’s where cloud control comes in. The best way to achieve it (and its associated peace of mind) is to establish a secure middleware layer for your cloud applications: a private Platform as a Service—or private PaaS—on top of a private cloud.
Quick Recipe for a Cloud Cake
Cloud computing offers elastic software services for enterprise customers. The great thing about virtualization is the “virtual” part: Abstracting software as a service for consumers (be they developers, IT leads, CIOs, or software users) shields them from the challenges of managing scalability. Without cloud virtualization, scaling apps to meet growing business needs requires resource, time, and hardware investment. In the cloud, that scalable capacity accommodation is there for the taking (and right price).
Clouds can (and should!) take several forms, depending on the business needs of the cloud-implementing organization. In a public cloud, service providers offer application- and data-hosting on their hardware in their datacenters. They price by storage space, number of virtual machines served, or even traffic. Alternatively, a private cloud is the self-hosted equivalent of the public cloud, with infrastructure and applications served from on-premise hardware. As its name suggests, a hybrid cloud combines both models, with some data and applications delivered locally, and some deployed to a third-party public cloud (as business needs dictate).
There are three “layers” to the cloud cake. The bottom layer is the Infrastructure-as-a-Service layer, or “IaaS.” In a public cloud model, cloud service providers like Amazon, Rackspace or HP Cloud Services provide this foundation for developers and enterprises to build upon. In a private cloud environment, the IaaS is hosted internally within an organization using solutions such as OpenStack, CloudStack, Eucalyptus, or OpenNebula. The top layer is Software as a Service, or “SaaS,” the application “frosting,” as it were, and that’s what’s ultimately served to the end users, be they internal employees or external customers.
