iot in the cloud.jpg

IoT in the Cloud: How to Build a Scalable Network [2018 Case Study]

See how you can bring your cloud IoT project to life
Last updated: May 29, 2018

Organisations are expected to spend around £53 billion on IoT (Internet of Things) services by 2020.

But why is the connected enterprise such a sought-after commodity? And what does your company need to do to operate at scale?

statista graph - size of global IoT services 2020

During our time as cloud specialists, we’ve been at the forefront of IoT innovation, helping our clients build reliable, future-ready IoT cloud platforms. In a recent engagement, we worked with a leading private healthcare provider, Care UK, to revolutionise traditional healthcare services - designing and deploying a network of wearable biometric devices.

Dubbed the WellWatch, this IoT project built upon the flexibility and elasticity of the Azure IoT suite, enabling the firm to offer life-saving insights as a service.

So, how exactly did we help Care UK reach their desired business outcomes and what are the key takeaways for your organisation?

Here’s our approach to building a scalable IoT platform and making the most of Azure cloud solutions.

New Call-to-action

Exceeding expectations with IoT in the Cloud

As well as looking to develop a new line of business, Care UK felt a social responsibility to provide a service that elevated the quality of healthcare they provide.

Their goal was to develop an IoT platform that offered 24/7/365 guardianship to vulnerable customers at a sustainable cost to the business.

WellWatch uses the device functionality of Huawei Watch 2, which include:

  • A gyroscope and accelerometer
  • A heartrate monitor
  • A temperature gauge

Combining these wearables with the Azure IoT suite, enables the firm to stream and analyse vital health telemetry in real-time, boosting their ability to respond to critical events such as falls or strokes.

An initial POC of our Azure IoT solution demonstrated that the network could alert emergency services to an event in less than 90 seconds, significantly improving response time.

Building a Flexible end-to-end Azure IoT Architecture

To ensure that WellWatch could provide a consistent service to hundreds of thousands of users, we built an end-to-end solution that optimised each component of the Azure IoT suite.

As with any IoT project, we focused on meeting Care UK’s expected business outcomes – in this case, driving innovation and offering a scalable customer service.  

To achieve this, our Azure IoT framework, EPiC, ingests gigabytes of biometric telemetry every second, using secure, bidirectional communications to transfer raw sensor data to an Azure IoT hub. The largest version of these hub servers can receive up to 300 million messages a day, with new servers deployable on-demand.

Once the data is collected, we use a number of Azure tools to perform advanced analytics:

  • Stream analytics: SQL-like rules to monitor and identify incoming data, looking for specific patterns and triggers. These include metrics like the user’s heartrate. If this is below an expected level, action is taken to notify the user, their next of kin or, in extreme circumstances, call an ambulance.
  • Machine learning: for more complex patterns (such as fall detection), we connect seamlessly with the Azure Machine Learning Studio. Using historical data to create statistical models, we can run predictive monitoring as stream analytics functions and respond proactively to any incidents.
  • Data storage and processing: with so much data to analyse, we don’t want to waste a single drop. Hot data is sent to an Azure Data Warehouse, where it can be analysed on a petabyte-scale using IoT visualisation tools. All other data ends up in a data lake, where technology such as Spark enables us to build a rich, detailed cohort of data and drive innovation within Care UK’s core business.

All of these components integrate natively, making it easy to test, integrate and scale your IoT platform at will. In Care UK’s case, this gave them the freedom to build a population-scale network, whilst managing the battery life of individual users.

Key Takeaways: making IoT Cloud Platforms work in your Organisation

The beauty of Microsoft’s Azure IoT suite is that the underlying architecture works for any IoT project. However, your specific use case will affect the way you approach the work. With only a quarter of IoT projects making it past the POC stage, it’s important to evaluate your IoT readiness before you begin:

1. Secure Stakeholder buy-in

If your C-suite isn’t onboard with your IoT ambitions, it’ll be difficult to secure the funding and support you need to make your project a success. Build your business case around tangible business outcomes and make sure you understand the true value proposition.

2. Identify Potential Roadblocks early on

Security, cost and lack of internal skills are all stumbling blocks on the road to IoT nirvana, but they’re not insurmountable. Azure’s flexible infrastructure makes it easy to rapidly test and deploy new IoT projects without racking up huge upfront costs.

3. Optimise Architecture and Monitor Performance

With edge computing and API integration changing the way we approach IoT, ensuring you build your project on a flexible and agile IoT platform is key. Technology such as the Azure IoT suite and our own EPiC framework enable you to gain real-time insights into system performance and identify ways to improve network efficiency.

To find out more about the work we’re doing with organisations like Care UK, check out our Azure IoT page.


iot guide

Topics: IoT, Cloud

KNOWLEDGE IS POWERGet weekly email insights from RedPixie

IT Partner Success Choose correctly and boost your value by 56% →
surface banner.jpg