Shortcomings in M2M development have been overcome with the advent of unified M2M service delivery platforms. This M2M architectural improvement places an additional layer in the network topography to house a central repository of task-specific functions needed to support otherwise disparate M2M systems. Sometimes classified as a “Platform as a Service” (PaaS), it plays an intermediary function to manage data exchange among any number of proprietary standalone M2M systems that were purpose-built for monitoring and controlling remote assets with their relevant applications, irrespective of the underlying communications network. Sitting between the application layer and device-level communications, it frees system designers to select devices from multiple sources, irrespective of embedded protocols at the lowest level of communication.
The key benefits of the M2M PaaS architecture over standalone M2M implementations are time and cost savings. The system is essentially agnostic, incorporating a set of pre-loaded depository of application building blocks that streamline initial deployment. Platforms can be programmed to accept most incoming data streams from M2M devices, and provide output data that conforms to the relevant applications program interface (API) available to the application developer. APIs ensure that any given application receives inputs in an understandable format, so that critical messages like failure reports can be routed and dealt with appropriately. This simplifies the task of upgrading M2M systems because the unique needs of applications are disaggregated from the core platform function – and, most importantly — from each M2M device and its unique communications protocols. By providing API specifications to developers, platform advocates point out that this architecture takes uncertainty out of the equation.
Platforms also add an element of scalability to M2M implementation. Beyond aggregating the streams of GPS and sensor data supplied by the various endpoints, platforms can be cost-effectively used to correlate the data streams with other cloud-stored enterprise data through the use of analytics. Now both real-time and historical M2M events can be rapidly leveraged to expose unseen correlations that can be used to drive informed business decisions in a timely fashion. (“Big Data” is discussed in the following section.)
The platform approach to M2M deployment is a means to “future proof” M2M applications and thereby reduce the risk of investing in them.
For more information, see: