10 IoT Technologies Every CIO Needs To Understand


The Internet of Things (IoT) will enable many innovative digital business opportunities, but to profit from them, CIOs must come to terms with many new and immature technologies and vendors. Ten technologies will be key to unlocking the full potential of the IoT.

IoT security

Many IoT security risks will threaten IoT devices, their platforms and operating systems, their communications, and even the systems to which they’re connected (using “things” as an attack channel). Novel threats such as “vampire” battery draining attacks are also emerging. No single technology or vendor will completely secure the IoT, CIOs must consider a broad portfolio of products and techniques.

 IoT analytics

The data collected by IoT devices will enable organizations to understand customer behavior, deliver new services, improve their products, and intercept and exploit transient business moments, so analytics tools are vital. But “things” will generate new types of data involving location and time series information that require new skills, algorithms and visualization tools. Also, the complexity of the IoT will demand complex distributed analytics with information processed in the “things” themselves, gateways and the cloud.

“Thing” management

Nontrivial “things” will require monitoring and management, e.g., to check their status, manage security certificates, commission and retire devices, update firmware, perform diagnostics, and collect information to forward to cloud services. In some industries, management will be a daunting challenge involving thousands, perhaps millions, of devices. Network technologies also impact management decisions, as cellular-connected things require additional platforms provided by the network operator.

 Low-power, short-range networks

Most “things” will be connected wirelessly, and an important cluster of wireless technologies is aimed at relatively short-range, low-power links in, for example, the smart home and office. At least 10 network technologies will compete in this space, differentiated by features such as range, battery life, bandwidth, cost connection density and so on. Examples of these technologies include ZigBee, Thread, Wi-Fi, Bluetooth, G.9959 and ANT. Networks are evolving, new technologies will emerge and established technologies will provide new features such as the forthcoming mesh Bluetooth standard.

IoT processors

The IoT doesn’t use conventional microprocessors but tends towards microcontrollers and is currently dominated by simple 8 bit devices which will only be overtaken by more sophisticated 32 bit chips around 2018. Selecting the right IoT processor sounds like a geeky topic but it’sis a key decision that defines many key features of an IoT product, e.g., whether it can be updated in the field, the cost of software development, whether data can be processed on the “thing” or must be handled externally, and what types of security can be implemented.

IoT operating systems

Traditional operating systems aren’t well-suited to IoT needs — they need sophisticated processors and lots of memory, and they consume too much power. As a result, a new generation of operating systems has evolved for resource-constrained IoT needs. At one end of the spectrum are variants of mainstream platforms such as Android, and at the other are minimal platforms such as RIOT, Yottos and Contiki, which have a minute memory footprint, as small as 5K bytes in some cases.

Low-Powered Wide-Area Networks (LPWANs)

IoT applications such as smart cities need a network where endpoints cost only a few dollars, services are inexpensive, bandwidth is low (from a few tens of bits to a few kilobits a second), coverage is nationwide and a battery can last many years. In 2016, the leaders in this space are proprietary technologies such as LoRa and Sigfox, but in the long term the battle will likely be won by new cellular systems such as NB-IoT. For now, CIOs may have to make tactical networking decisions and plan to update their hardware when more-standard solutions become available.

Event stream processing


Billions of connected devices will generate immense quantities of data; tens of thousands of events a second are common in some industries and millions of events a second are found in areas such as telecommunication. Conventional “store then process” systems can’t handle these situations, so CIOs must consider novel, highly parallel technologies such as distributed stream computing platforms.

IoT platforms

One way to deliver IoT solutions more rapidly is to build them on a IoT platform — a software product that bundles together many of the functions needed in an IoT system. These platforms provide a spectrum of features, from low-level device management and control through data acquisition to application delivery, analytics and bridges to enterprise systems such as ERP. However, the reduced time to market must be set against a relatively high cost per “thing” and the fact that many of these products are proprietary.

IoT standards and ecosystems

Ecosystems and standards aren’t precisely technologies, but we include them here because most eventually appear as APIs that will enable devices and systems to interoperate and communicate. Low-level communications standards such as MQTT and CoAP are well defined, but at higher levels there is a chaotic mix of overlapping standards bodies and competing ecosystems in areas such as the smart home. Many “standards” groups have agendas to promote specific technologies or architectures.