The IoT: The network can make it or break it
In a previous article, we introduced the Internet of Things (IoT). The IoT is emerging thanks to the collaboration ― as well as the convergence ― of communications, information technology, and consumer electronics industries.
Although today’s media focus is mainly on devices and applications, it has to be well understood that the machine-to-machine (M2M) communications network can “make or break” this next step in the evolution of the Internet.
A CHALLENGING INFORMATION & COMMUNICATIONS TECHNOLOGY ENVIRONMENT
In a recent FutureScape report, IDC predicts that:
- Within the next 5 years, more than 90% of all IoT data will be hosted on cloud-based service provider platforms.
- Within 3 years, 50% of IT networks will transition from having excess capacity to handle the additional IoT devices to being network constrained with nearly 10% of sites being overwhelmed.
- By 2018, 40% of IoT-created data will be stored, processed, analyzed, and acted upon close to, or at the edge, of the network.
- Within 2 years, 90% of all IT networks will have an IoT-based security breach, although many will be considered "inconveniences."
- By 2017, 90% of datacenter and enterprise systems management will rapidly adopt new business models to manage non-traditional infrastructure and BYOD device categories.
Current IoT users, service providers and industry analysts all emphasize the need to reduce complexity, to optimize the end-to-end architecture and to provide a better experience for all stakeholders:
- Business Insider: “The IoT lacks a common set of standards and technologies that would allow for compatibility and ease-of-use. There are currently few standards (or regulations) for what is needed to run an IoT device. Consortia that group together global industrial, tech, and electronics companies are involved in an effort to standardize the IoT and solve the most pressing security concerns.”
- Harbor Research: “Current IT and telecom technologists are operating with outdated models of data, networking and information management that were conceived in the mainframe and client-server eras and cannot serve the needs of a truly connected world. ‘Smart Systems’ should automatically be understood as ‘real-time networked information and computation’, but it isn’t.”
- ABI Research: “… a key challenge for the industry remains the complexity of developing, deploying, and managing M2M applications ... This is a challenge both for mobile network operators that are trying to offer profitable services tailored to the M2M market, as well as for application developers and service providers that are trying to reduce costs, speed time to market, and simplify robust application deployments.”
- Strategy Analytics: “For many years M2M was held back by the lack of a low cost, global access medium, the fragmented nature of the ecosystem, the lack of any single killer application driving demand and the complex nature of M2M solutions leading to high-cost development and systems integration. With the shift to IP service platforms and ubiquitous connectivity, M2M is now becoming a serious means of facilitating business transformation.”
THE VALUE OF THE NETWORK
It should be clear that the Internet of Things (r)evolution can only succeed when it’s supported by a network that allows scalable deployment, secure delivery, cost-effective operations, and fast time to market of new applications.
Basically, in an IoT network there are 2 opposite-directioned data flows:
- A device-outbound stream, through which sensor and device data are delivered to the network, the data center and the respective applications.
- A device-inbound flow via which actuation, control and management information is delivered to M2M devices and gateways.
In most cases, the device-to-application data will far outweigh the application-to-device traffic. But at times, there will be significant throughput in the latter direction, too. In the upstream direction, the traffic may be very application specific ― sometimes the flow will be continuous and sometimes bursty. Traffic modeling by Bell Labs has revealed that M2M applications may consume up to 67% of computing resources in the radio network controller. Access channel capacity does not appear to pose a problem, nor does data volume.
The IoT growth figures cited above are impressive. The time is now for anticipating massive take-up of devices, applications, traffic, and profile and usage data.
This can only be achieved through a network and platform infrastructure that is scalable by design. A network that includes overload protection mechanisms at the RAN and in the core network, combined with application-level control to enable more efficient use of network resources.
As many of the capabilities will be implemented in the cloud, SDN/NFV-based networks will provide carriers and enterprises with the necessary means to cope with and manage the growing number of devices and applications, and the IP traffic they generate.
As some devices may deeply be embedded in 3rd party infrastructure (such as industrial equipment, transportation containers, and cars) and may have no registered owner at all, remote management capabilities become extremely important.
The adoption of device management and service automation capabilities will also help the M2M service providers accelerate device and application onboarding while reducing operating costs. These capabilities streamline tasks such as:
- Remote device activation and bootstrapping
- Device configuration
- Firmware upgrades
- Application lifecycle management
The legacy M2M market has been fragmented by divergent protocols and custom applications. This fragmentation makes it difficult to handle devices consistently and develop solutions that can apply to more than one vertical market.
M2M service providers need to stop deploying separate stovepipes for different applications, and work towards an “any device, any app, any network” model. An end-to-end network architecture with a common set of service capabilities, standardized interfaces and open APIs should help them to reduce investments, facilitate partnerships, and speed up time-to-market.
Deployment of a horizontal M2M control and management platform will allow them to abstract devices and applications from the underlying access networks and technologies, which will in turn result in reduced development effort, lower operational expenses, and a better customer experience.
Cloud technologies such as software-defined networks (SDN), network function virtualization (NFV), and datacenter hosted services will facilitate initial deployment and enable smooth growth of IoT applications. SDN will help transmit and process the data generated by an explosive number of IoT endpoints without further putting the network under pressure, while capabilities such as service chaining, dynamic load management, and bandwidth-on-demand will make service providers more agile.
Infrastructure virtualization is a means to derisk IoT investments, to support innovation, and to keep operational expenses low. Recent business modeling by Bell Labs has shown that virtualization of the enhanced packed core may lead to up to 40% savings in total cost of ownership (TCO) for M2M services.
In a world where security and privacy are playing a more and more prominent role, functions such as authentication, authorization, encryption, and data protection score high on the IoT checklist.
According to ABI Research, “Most M2M applications are lacking the basic security requirements that have been a de-facto standard for information and communication technologies elsewhere. If not addressed sooner, this weak link could throttle the successful adoption of M2M in healthcare, industrial installations, and consumer homes.”
Security will have to be supported by the network cloud, and not only by the devices. Data centers will need to ensure that information is available at all times, and that the big data as well as the connections over which they are transported are protected against loss and unauthorized access. As communication service providers are already seen as “trusted partners”, there is an opportunity for them to build on this relationship in the development and the commercialization of IoT services.
In order to achieve the “any device, any app, any network” objective mentioned above, devices and applications need to get abstraction from underlying access networks and technologies. This model can only work when there is maximal interoperability between devices, platforms, data formats, protocols and applications.
Furthermore, because many M2M devices are often characterized by very small power, memory, and processor footprints, communication and management protocols have to be simple and lightweight.
The development and adoption of standards, such as oneM2M, will harmonize device interactions, simplify integration and create economies of scale. Standardization will also make it easier for individual stakeholders to partner and interwork with each other’s components, networks and services.
Big data handling
BI Intelligence forecasts that 40,000 Exabytes of data will generated globally by 2020. When M2M platforms open up their data, new applications can leverage intelligence in objects and in the cloud. Huge amounts of data points may be correlated and aggregated into analytics, making networks and applications smarter and giving users full control over different domains such as security, multi modal transportation, retail, e-government, social statistics, environmental measurements, and senior or children assistance services.
The generation of these analytics and their exploitation needs big data enabled IT infrastructure. And big data applications require a superior network: backhaul for all connected objects, open data center interconnections, a cloud SDN distributed architecture, etc.
EVERY SUCCESS HAS ITS NETWORK
The IoT emerges as an unprecedented business opportunity for many players in the communications, information technology and consumer electronics industries. And a means for service providers, enterprises, and verticals to enhance productivity, achieve cost savings, and/or transform their business.
Alcatel-Lucent is a leading IP networking, ultra-broadband access and cloud specialist. We have the right assets to partner with M2M service providers, enterprises, utility providers, and public administrations in building the infrastructure for the Internet of Things, providing a better customer experience at a lower TCO, while helping them move up the M2M value chain.
-  “M2M Dream Challenged by Alarming Security Concerns”, ABI Research, 2013
-  “The Interoperability Enabler for the Entire M2M and IoT Ecosystem”, oneM2M white paper, 2015
-  “The Internet of Everything: 2015”, BI Intelligence, December 2014
To contact the author or request additional information, please send an email to firstname.lastname@example.org.