I have no guru-type pretentions, but last year in an article entitled “M2M in the Cloud: a logical development” I stuck my neck out and wrote this opening: “Mainstream apps and business processes are moving into the cloud, but right now MSM runs on proprietary platforms, which blocks any move into this cost-effective ICT environment.”

I also referenced a short ETSI (News - Alert) article on standards, the conclusion of which was the need to replace the current vertical, stovepipe approach with a horizontal ICT-type model that enables system elements to be shared. And later on, at ITEXPO (News - Alert), I was told that we could expect to see real progress from ETSI on standards in 1Q 2011.  Well, the title indicates that progress has been realized and it looks (advance apology for the cliché) as if we’ve come to a significant M2M inflection point.

The new environment

ETSI’s standardization process started two years ago, and now there is an agreement on a high-level system architecture as well as the requisite service capabilities. As illustrated, the architecture, which is based on the existing communications standards of flat, all-IP next-gen networks, has extended M2M capabilities in the core networks as well as a separate device domain.

The core network provides: IP connectivity; service and network control; interconnection (with other networks); and roaming. The M2M service capabilities provide: functions that can be shared by different applications; exposure of functionality via open interfaces; use of core network functionality; and simplified, optimized application development and deployment.

The italicized words are particularly significant as this development will allow M2M applications to migrate from the current vertical model and adopt the horizontal cloud-computing model of today’s enterprise environment. 

The inflection point

That was a very quick take on tomorrow’s integrated environment -- quick because the really interesting development that will accelerate M2M’s momentum is the new device domain, which has three elements: the devices, an M2M local area network (e.g. ZigBee), and an M2M gateway. 

The devices can connect directly to the network domain and they continue to perform procedures such as registration and authentication, but they can also employ the service capabilities of the network. Alternatively, the devices can connect to the network via the gateway, which provides the same procedures on behalf of the devices. This means that the gateway can also execute applications on behalf of the network domain. ETSI’s gateway is therefore a standards-based M2M platform and it will only be a matter of time before today’s proprietary middleware products are replaced. 

The inflection point comes from the fact that this development will allow application developers to leverage the functionality of intelligent wireless modules, i.e. their ability to process data at the device level, and in turn this will accelerate the deployment of applications that run in the device domain over a local M2M area network. In this scenario the wide area core network would only be used to transmit information to a central facility so that it could be displayed, e.g. on a Web portal.

The implications

Let’s take a quick trip down memory lane. Once upon a time, data processing was conducted in vertical hierarchy: dumb terminals communicated with IBM (News - Alert) mainframes and there was no local processing. LANs flattened the architecture and PCs, which are intelligent terminals, enabled the client-server model and the rest, as they say, is ICT history.

The analogy with the M2M is clear. Today’s model employs embedded modules (clients) transmitting data over a network to an M2M platform (server) that converts the data into information. These modules perform relatively simple tasks, but when powerful processors are embedded local intelligence can be used to employ analytics that support capabilities such as making real-time decisions, reducing data traffic and identifying critical events.  

In the new model, the applications can run in the device domain over an M2M LAN and the gateway concept can unleash the considerable power of advanced processors such as the quad-band, dual-processor ARM9, which is embedded in Telit’s GE863-PRO3 wireless module. Dual-processor chips, as the term implies, can perform two tasks simultaneously, e.g. one processor allows software applications to take full control of the CPU while the other handles time-critical communication tasks.

Therefore, one doesn’t need to be a guru to see wireless modules rising up the value chain in line with increased functionality as well as multi-band air interfaces, which provide global connectivity. And of course that development is facilitated by Intel’s (News - Alert) M2M activities, so maybe some day we’ll see modules branded with “Intel Inside.”

Conclusions

M2M has enjoyed a remarkable success, and it came through the recent economic recession in good shape, but open standards are needed otherwise M2M will never realize its full potential. The new M2M environment, with its flexible, synergistic combination of service capabilities in the core network and the device domain represents a significant step towards the realization of that objective. The flat architecture and the ability to share components will accelerate the development of new, as yet unforeseen applications and also facilitate the integration of M2M apps into the mainstream, cloud-centric ICT environment, thereby leveraging the initial investment.