What the Internet of Things Is Missing
What the Internet of Things Is Missing
The IoT has limitless potential, but it has been held back by an inability to properly scale. Can automation change this?
Join the DZone community and get the full member experience.Join For Free
For what feels like a lifetime, industry analysts and experts have predicted this will be the year the Internet of Things finally takes off, both within industry and for consumers. But, as if by clockwork, each year the "guaranteed" IoT explosion passes, the mass adoption and exploitation of the Internet of Things having failed to materialize. So, what is preventing the IoT from gaining any meaningful traction?
The IoT Vision
The IoT vision is almost utopian in its promise: a connected, intelligent world, offering a simpler, more convenient future. Our houses will have smart thermostats that know precisely how warm we like our rooms, speakers that understand our taste in music and refrigerators that realize when we're running low on the essentials and act to replenish themselves. Our highways will be populated by sensors, bringing congestion to an end; our streets will be safeguarded by interconnected surveillance. This will all be made possible via the communication of a global set of devices.
What potential this vision has! Gartner predicts that, by 2020, more than 20 billion 'things' will be connected to the Internet. This connectivity could touch all businesses aiming at near real-time interactions with consumers, ranging across all sectors: insurance with accident sensors, health with biological or physiological sensors, manufacturing with maintenance sensors, marketing with location and environmental sensors, utilities with smart meters... The list goes on.
There have been significant barriers between the realities of the present and the promised land of a truly connected world. A lack of defined industry standards is one such hurdle. To date, the IoT hasn't been populated with devices designed around the same building standards, or even with unified security protocols.
Similarly, the vast differences between the functions and types of data created and used by different 'things' must be reconciled. The Internet of Things is so diverse, it can consist of just about anything: cars, watches, doorbells-even egg trays. As such, connected devices struggle to simply talk to each other organically, and each device requires its own application, rather than everything being controlled from a single central hub.
From a consumer's point of view, this lack of central control is holding back adoption. For IT departments working on IoT-related projects, the huge volume of connected devices poses different challenges, including security, communication, processing, storage and complexity. As a result, Gartner estimates that in 2018, more than 75% of IoT-related projects will have lasted twice as long as expected.
Then there is the challenge of harnessing and effectively using the nearly limitless amount of data the IoT will produce. A 2016 Network World article declared that a single autonomous vehicle will create four terabytes of data per hour. IoT devices are active data collectors and reporters, but sorting through the volumes of data to make meaning of it is still a challenge. Artificial intelligence (AI) and machine learning offer the potential to help interpret large datasets, but those engines fizzle if off-script datasets erupt on the scene, and can become dangerous when the AI is responsible for human life (e.g. self-driving cars) and encounters datasets outside its training.
Automation, the IoT and the Modern Software Factory
In the digital era, automation is pivotal to the modern business, to digital transformation and to the eventual success of the IoT. Attempting to manually push updates, security fixes or new features to IoT devices would be a fool's game. To make it really feasible, we need more than hardware and connectivity: we need software that makes it possible to run at scale. The Internet of Things relies upon continuous delivery and continuous deployment practices which, as we know, are only possible at scale with automation. The DevOps approach of writing small, iterative builds consisting of frequent batch releases of code are well suited to the IoT and vital for ensuring security and maintaining performance. At the same time, it is necessary to have a solution that can introduce new levels of scalability and speed to the orchestration of data flows within the organizations that provide-and use-interconnected products and services.
The Internet of Things is closer than it's ever been to living up to its potential. Although we may be some way from universal consumer adoption, the right automation solutions are key to facilitating the necessary communication, scalability, and speed to finally connect everything.
Published at DZone with permission of Scott Willson , DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.