This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.
M2M As An Integral Part Of The Enterprise
M2M solutions deliver significant benefits, as evidenced by the track record of this innovative industry, but the great majority of the wireless M2M solutions are realized in vertical, so-called silo architectures. In addition, proprietary technology is employed because there was and still is a dearth of standards.
This is in stark contrast to the enterprise model, which is horizontal and based on standards. Therefore M2M applications such as fleet and asset management and mainstream business apps like ERP and CRM operate in environments that are significantly different from enterprise environments, both technically and culturally, and this represents a significant challenge when trying to integrate M2M with the enterprise.
Ideally, integration would be based on M2M solutions that are standards-based, open and cloud-centric: see "M2M: Looking Further Ahead." However, this development only facilitates integration: it does not enable the bidirectional seamless transfer of information between M2M apps and business processes.
M2M solutions emerged more than 20 years ago, although the term was not used at that time. They were literally machine-to-machine, being deployed on factory floors and in vending machines as well as a few specialized industries, e.g. oil and gas. Communication was wireline and the data was normally integrated into business processes. However, in recent years, the M2M development action and innovation has increasingly been wireless--short-range RF and wide area cellular, with solutions proprietary and silo-ed.
A recent survey conducted by Axeda indicated that companies having physically static assets that generate M2M data on events or parameters are increasingly likely to use wireless because of the rapidly falling cost of cellular versus other alternatives. Fixed lines often need new labor-intensive installation activity and have a requirement to route data through on-site networks. Cellular connections enable direct wireless connections to machines:from the asset to the cellular network to the application platform. There is no need to connect to the corporate network: no need to involve local IT.
The Axeda survey also showed that the majority of the early adopters who participated are planning to integrate their connected products into the enterprise so that the data can be shared with other departments. Around 11% have already integrated M2M data, and another 67% are interested, planning or scheduling integration with ERP and CRM systems in order to: optimize field service capabilities; enhance customer service; and automate asset management and configuration management.
What are the Key Issues?
Integration is relatively easy when data is transmitted over an enterprise LAN; communication over a wide area cellular network is radically different. For example, there is a long, multi-vendor value chain, which includes the mobile network operator, and there are operational issues, e.g., the SIM cards have to be deployed, activated and managed. It's not rocket science, as evidenced by the proliferation of M2M apps, but it does require the specialist knowhow and experience of a solution provider.
Moreover M2M is a fragmented industry, which breaks down into nine main sectors and numerous sub-sectors (see the "M2M World of Connected Services" map on Beecham Research's site for details).
On the other side of the equation, there is an equally complex set of integration challenges, e.g., whether to use SOA (service oriented architecture) stacks or a lightweight standalone ESB (Enterprise Service Bus, a software model used for designing and implementing the interaction and communication between mutually interacting software applications in a service-oriented architecture).
Systems integration therefore requires specialist knowledge and experience in both the M2M and enterprise environments, and this is hard to acquire. The big SI players tend to see M2M as an industry that operates on the dark side of the moon, and regular M2M vendors are way out of their comfort zone when it comes to the enterprise back-office systems. That's the main reason why we remain at the early adopter stage.
As Robin Duke-Woolley, CEO of Beecham Research puts it: "M2M is now on the radar of CIOs. Integration with ERP and CRM is very much on their agendas, with increasing interest in how such data can strengthen the competitive edge of their companies in their own markets. As part of that, there is also growing interest in data analytics to store and mine the data from remote devices in order to gain new intelligence about customer needs and new market trends."
Feet in Both Camps
Axeda has the requisite set of M2M and enterprise skills, which they position as the "Internet of Corporate Things": a B2B subset of the notion of the Internet of Things. Their offering is predicated on the management of the physical products that an enterprise delivers to its customers, and the management of that company's corporate assets, including facilities and plants, operational equipment, vehicles and goods being delivered. And it begins with the fundamental understanding that supporting customers' use of a company's products requires connectivity, monitoring, remote service, usage analysis, ERP/CRM integration and value-added services.
The offering also includes managing internal corporate assets and operational infrastructure, which requires an understanding of the complex web of interconnectivity between assets, places, people and information. Companies need to integrate M2M data into their mainstream applications and processes in order to leverage the functionality of their connected products and assets, and thereby create new applications and business models that differentiate their offer.
This schematic is a neat illustration of the way that data emanating from various M2M solutions can be leveraged when it is integrated into an enterprise's CRM/ERP systems. Alarms, for example, can generate a CRM ticket or case, which improves customer and field service. Data can also be sent to a billing or supply chain management system in order to eliminate the mistakes that can come from manual processing.
Axeda's platform includes application services and data management, as well as an integration framework that employs standards-based message queue technology to enable integration with enterprise systems, including ERP (e.g., Oracle and SAP), CRM (e.g., Salesforce.com), plus billing and data warehouses.
Connectivity services, software agents, and toolkits are employed to establish connectivity between devices/assets and the Axeda platform, and all major media types can be used: cellular, the Internet, Wi-Fi, or satellite. Connectivity to legacy devices/assets, i.e., those that have been deployed and that run in a silo solution, is provided in different ways. For example, agent software can be installed on or near the devices. Alternatively, the proprietary TCP/UDP messages can be translated into the company's data model. This is done on the fly.
Obviously this process must not interfere with the legacy solution. This is accomplished by the rules engine in the cloud that decides what to do with the data. If the data is for the legacy app, then it will be forwarded to the regular application server: no rule is applied. Data destined for integration can have different rules; for example, if the device data is indicating an abnormal condition, then it will be forwarded to the relevant mainstream app because it's data that they may need to see immediately. According to Axeda, this is a key requirement because if a pure message broker is used then all the M2M will go to the backend systems, which will be overwhelmed.
Next page: More players
ILS Technology is another player having both M2M and enterprise skill sets. The company is 10 years old, but says that is has been active in M2M for 30 years, an apparent contradiction that comes from the fact that ILS is a spin-off from IBM. The founders and most of the team were in Big Blue's factory automation department, where they did last-mile integration, e.g., remote access to the machines employed in semiconductor fabrication plants.
Along the way, the company has developed a comprehensive portfolio of application transports and connectors for the majority of mainstream enterprise applications: those of IBM, Microsoft, Oracle, SAP, etc. They are deployed in the enterprise gateway (see the following schematic).
Connectivity to devices is enabled in three ways: (1) by installing gateway software on the device; (2) use an Asset API when the device can't support the software; and (3) create a custom proxy to integrate devices to the M2M platform. Number (3) is the preferred approach for devices that are already deployed. As illustrated, the software is located in the asset gateway. Systems integrators and IT departments can employ these products in order to simplify and speed up integration, and ILS also provides support services.
The "deviceWISE" application platform, which is located in the cloud, is the bridge between the two environments. It enables direct links to be established between intelligent devices and the enterprise environment.
This schematic illustrates ILS's M2M application framework, and it summarizes the deployment functionality. Key usage functions include direct connectivity to SAP, SQL databases and enterprise service buses.
A really neat feature is the ability to leverage most existing M2M investments without a forced need to migrate. This is done by copying the data stream from the devices and sending it to the enterprise gateway, thereby leaving the legacy application in place.
"Everyware Cloud": an M2M Integration Platform
The term EveryWare Cloud comes from Eurotech, and like ILS, Eurotech was doing M2M integration long before the term was coined. For example, Eurotech developed solutions for the oil and gas industry more than 15 years ago, as well as numerous projects that employed M2M technical building blocks that connected intelligent devices and sensors in the field with business applications. In addition, the company worked closely with IBM on the direct connection of their devices into IBM Websphere, and this led to the development and open sourcing of MQTT (Message Queue Telemetry Transport).
EveryWare Cloud functions in a similar way to Gartner's concept of Integration Platform as a Service (iPaaS). iPaaS is basically a platform that enables integration within the cloud and between the cloud and enterprise. It allows users to develop integration flows that connect applications residing in the cloud or on-premise and then deploy them without installing or managing any hardware or middleware.
Right now this is a somewhat controversial topic, but Eurotech found that it was in line with their thinking, and their solution follows the general concepts of iPaaS (see the next section below). However, the design focus is on integrating distributed systems into the enterprise IT world through standard protocols and APIs. This allowed a message-based "Enterprise Service Bus for Machines" architecture to be incorporated, and that allows easy integration of different device data systems and applications. It also enabled device data management and device lifecycle management, as shown in the schematic below.
This is a comprehensive solution that provides the functional elements needed to perform message transformation, message routing, protocol conversions, data normalization, service virtualization, tracking, accounting, administration plus life-cycle management of the distributed devices. This means that EveryWare Cloud delivers additional functionality to the enterprise environment. It allows the M2M infrastructure to be seen from the IT perspective just like another enterprise application, and it also allows interaction with the M2M infrastructure in IT-centric ways.
An important feature is the use of device/protocol-specific adaptors to retrofit legacy solutions. They are created using modular software building blocks, e.g., Modbus and CAN bus on the device side. On the enterprise IT side, there are generic adapters, like REST APIs for device data management as well as device management, and different standard ways of retrieving device data. In addition there are application-specific adaptors for email, SMS, TingWorks, Twitter and soon Salesforce.
The Everyware Cloud transforms bits of data at the edge of the distributed device network into valuable actionable information in the business user's hands. It decouples data producers (sensors and devices) from data consumers (applications). A lightweight protocol, MQTT, is used to communicate with the devices; standard protocols and APIs are used to communicate with the enterprise applications.
The business logic on the device side is done using Java, facilitated by partnering with Hitachi and Oracle. Java corresponds not only with Enterprise IT approaches and best practice, but also solves a massive "resource scalability" problem associated with the IoT. In addition, the inclusion of an OSGi (Open Services Gateway initiative) layer in the software framework allows enterprises to add, amend and drop M2M services in line with changing requirements.
iPaaS: Integration for the Cloud Era
The cloud is changing the IT landscape, but as companies move more and more apps into the cloud they start hitting an ironic problem. Having spent years integrating their enterprise applications in order to abolish silos of information, they now find that "cloud silos" have emerged. The answer, according to Gartner and various vendors, is iPaaS, a cloud integration platform that enables secure connectivity to SaaS and cloud services.
One vendor having an improbable name, MuleSoft, pinpoints the problem as being the emergence of application platform elements without corresponding integration solutions. Therefore there is a need to link, integrate and orchestrate between these disparate components.
Gartner sees iPaaS emerging as the next-generation integration technology, gradually replacing traditional forms of integration middleware. This indicates the need for a seamless transition from on-premises IT architectures toward a hybrid model in which architectures coexist and interoperate with public cloud-based architectures.
The emerging IT software stack has three layers. Infrastructure as a Service (IaaS), a virtualized environment on which systems can be deployed; Platform as a Service (PaaS), an application infrastructure; and Software as a Service (SaaS), which provides the user-facing applications.
CloudHub, shown in the PaaS layer of the schematic below, is MuleSoft's iPaaS offer, a fully-featured solution that allows customers to take advantage of the economics and elasticity of the cloud for their integration infrastructure.
M2M solutions and IT systems occupy separate environments and they have evolved in separate cultures. M2M transmits real-time data from small devices and processes into actionable information. IT systems employ batch processing, the old mainframe computing model, and they deliver results after the fact.
Cloud computing is the development that is driving convergence--bringing them together in order to create a unified environment that leverages the M2M investment and extends the functionality of the business applications. In a nutshell, and here I am referencing a white paper from Harbor Research: When M2M inputs are integrated into systems that connect people and processes, the result is real-time knowledge that enables collective awareness, efficiency and better decision making.