Gavin Payne is a principal architect for Coeo, a SQL Server professional services company, and a Microsoft Certified Architect and Microsoft Certified Master. His role is to guide and lead organisations through data platform transformation and cloud adoption programmes.
Whatever your focus as an IT professional the interconnected world of the Internet of Things, known as the IoT, will bring more innovation to a world that’s already providing new technical capabilities on a regular basis. A world where IoT sensors will produce new sources of data that allows automated judgement and artificial intelligence to come one step nearer to our daily lives.
The IoT era will allow almost anything we want to be connected to the Internet to tell us what it’s doing or what’s happening around it. The sensor hardware needed will soon become so cheap it’ll be built into everything from thermostats to coffee machines to water pumps to factory conveyor belts. The cost, size and simplicity of sensors will see them deployed on a scale that’s never been possible and that leaves traditional IT monitoring systems quickly looking oversized and dated.
To describe an IoT solution, imagine driving to a city for the first time and wanting to park as close but as cheaply as possible to your meeting. Your smartphone will use the latest information from the city’s car parks to give you directions to a car park that currently has the cheapest hourly rate and the most free spaces - while being the nearest to your meeting. We’re already familiar with satellite navigation apps using real-time traffic information to plan the route to our destination but imagine a world where your smart phone decides what the best destination for you is - never mind how you get there.
Providing the IT services IoT solutions will need
When we start thinking about mobile devices making automated decisions using real-time sensor data we’d be forgiven for wondering how it’s all going to be possible – especially at a time when traditional IT is already being made to work harder and harder. The remainder of my article looks at how we might go about meeting three of the demands IoT solutions will have of IT.
1. Capable and scalable platforms
IoT solutions have the ability to receive, analyse and forward large amounts of data in near-real-time 24/7. When organisations need to process those scales of data, whether it’s the temperature of a room or the flow rate of traffic, they’re going to need scalable access to compute and storage to power the Big Data analytics tools used to find the gems amongst the mountains of data. Meeting these platform requirements are what cloud services excel at – scalable and elastic capacity and mass data storage that’s all deployable on demand.
In the initial days of IoT adoption we might be tempted to re-use existing on-premises IT to host IoT solutions as they do their data crunching. This might feel a good way of meeting a short term need when scalability and capability expectations are low but when the IoT becomes the best way to learn about the outside world then our focus will quickly shift to extending platform capabilities without time to worry about platform scalability. This is the point where public cloud services, whether Infrastructure as a Service or Platform as a Service, will allow IT professionals to focus on maintaining and enhancing IoT platform capabilities and not focusing on meeting a platform’s scalability requirements.
2. Global Scale Networking
Gartner estimate there will be 26 billion IoT sensors in use by 2020 but today we’re struggling to grow the Internet because of IP address shortages and complex traffic routing. The IoT works around these problems by making IPv6, the next generation of the Internet’s networking, its foundation. Not only will our application platform and data centre networks need to support this new networking standard but they’ll also need the capability to interface directly or indirectly with thousands or millions of sensors using new IPv6 addressing. IT teams will need to know how to manage connectivity with sensors wherever they are in the world, how to troubleshoot the loss of data from sensor to end user and how to provision enough low latency bandwidth to keep data flowing. Although cloud service providers will help with global scale networking, IT professionals will need to know how to implement and troubleshoot global scale sensor networks that use IPv6.
3. Device Security
The mention of networks sending and receiving data leads into the third area: device security. IoT solutions need to be ready to receive sensitive data from sensors they need to know are trustworthy and credible - yet security is still one of the challenges in the development of embedded technologies. Device authentication and traffic encryption are expectations the web era has made us all expect of any IT solution thanks to the ease of deploying SSL. Today we take it for granted that a secure web site is both trusted, verified and safe enough to share personal and financial information with. If we have sensors sending equally as sensitive information – perhaps medical data or information that implies whether we’re at home or not – we’re naturally going to expect the same level of assurance.
It’s unlikely anything we do today in our own data centres will natively scale to control access and secure data from millions of sensors that will one day cost less than $5 to produce. It’s also unlikely such cheap devices will initially offer the same grade of encryption we’re used to seeing on much higher powered smart phones, tablets and web services. Instead, we’re likely to see mobile device management solutions grow from supporting and securing a few thousand smart phones and tablets to millions of cheap sensors. Needing such technologies will see a shift in the IT security skills needed to deploy these solutions. IT professionals who are today familiar with managing computers and servers will need to transition to securing and re-risking sensors to meet the security expectations of tomorrow’s end users.
Having discussed just three points we might be forgiven to thinking this is a natural evolution of existing IT systems, their existing problems and their existing capabilities. Conceptually perhaps, but the ease of deploying millions of sensors that will depend on new compute and storage platforms, network technologies and security technologies shows how quickly IT professionals will need to adapt to supporting the Internet of Things.