What is a Device Cloud (architecturally speaking)?
A cloud can best be illustrated as a stack of layers that work in concert to provide the end-user the desired result. If you conduct a search on “cloud computing” you will find many different graphical representations. There are many interpretations. However, there is a common thread in all the different representations of the cloud. You will consistently find the presence of 3 distinct layers: Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). For the purposes of this topic we will use the following illustration:
Although the exact structure of any cloud stack will vary from implementation to implementation, this illustration provides a generalized guide. Each layer of the stack is further segmented by additional layers that represent the core functions. SaaS typically includes an application, data and the presentation or User Interface. The PaaS includes application programming interfaces or API’s, Middleware, and Integration of services. The IaaS includes the data center or facilities, wide area network access and hardware and related hardware management software.
As previously mentioned, the adoption of the internet has largely been driven by human-oriented applications. Therefore, the generalized cloud stack architecture is representative and has quite obviously and subsequently been influenced by the “people network.” But what about the “Internet of Things”, or devices? Does the traditional cloud stack work effectively for device or M2M networks? The answer is: Partially.
For the cloud architecture to be effective for delivery of services for M2M networks, we must extend the traditional cloud stack to include the local device network stack. In the device network, a stack much like the layers that exist in the cloud stack, exists.
The local device network stack includes the end-device, which can be a sensor or an adaptor, the local network (connectivity), and a method for aggregating and/or transforming data; and connecting to the Internet. A Gateway is a common way to provide the latter functions.
The Internet of Things will largely be enabled through this architecture. The primary reason is cost. Through the virtualization of many of the layers, but not all, within the stack; cost can be driven down to levels not achievable with enterprise application architectures.
So what is the Device Cloud? Perspective is key. It extends the traditional cloud stack to include the device network stack. Thinking of it this way can lead you to many possible conclusions regarding the creation and delivery of cloud “services.” More on that next time…