Skip to main content

Expanding the frontiers of connectivity: Scaling your architecture for the Internet of Things


In these scenarios, the challenge of adopting an IoT strategy is partly about collecting the data from these devices and processing it, and partly about making that resulting data actionable and linking it to the operational systems that help you know your customer or your business.

Many organisations are not set up to do either of these well at scale, and IoT technology domain creates some additional challenges around connectivity, reliability, and security that warrant an evolved approach to IoT architecture.

IoT is changing the computer hardware model we’ve had for over 40 years. Developers and architects are used to building software systems across two physical tiers. For IoT, there’s third hardware layer. This is an emerging concept, spearheaded by research at Princeton University, and it’s called the Fog Layer or the Edge Layer and is what devices connect to before the back-end systems - the server.

The new Edge Layer

The Edge Layer connects devices locally, and manages data collection and the connection to the servers. It filters data; protects against connectivity problems; enables site level orchestration across devices from different vendors using different protocols; and enables the server/cloud application to be agnostic to the device implementation it controls.

The Edge Layer has three main components in a typical IoT deployment:

  • The device or sensor itself: In IoT this is the client that generates data and/or receives commands to execute.
  • Usually a gateway to enable access to the internet or private network: Typically these gateways speak a proprietary protocol between the connected devices and then allow connectivity through the gateway using a standard protocol such as HTTP 
  • The Edge Controller: Responsible for connecting to all the gateways and independent devices in a physical location. The Edge Controller collects and collates data from all the devices, transmits data, and accepts commands from the server to execute across some or all devices  

The server layer is similar to what we see in other client server architectures, such as mobile:

  • The API layer: Leveraged by the IoT architecture to connect to the server layer. This provides consistency, control, governance, security and an analytics-based model for connecting external sites
  • Device management: Responsible for knowing what devices are within the IoT network and sometimes part of the authentication chain
  • The data plane: Providing event streaming, transformation orchestration and connectivity to the applications and systems so that it can use the data coming from the Edge Layer
  • SaaS applications, enterprise applications, big data and business intelligence: Consuming the data coming from the Edge Layer and making it actionable through analytics, dashboards, and application processing 

Companies getting the most out of connected devices are identifying value niches that can benefit from IoT technology.

Creating new consumer experiences: connecting with pets

There are many IoT scenarios that do not require an Edge Layer, but can leverage the server layer architecture to connect devices. Company i4C provides a pet collar that behaves like a FitBit for dogs. Creating this collar presented challenges. Unlike FitBit users, dogs don’t have smartphones, through which the collar can connect to the internet. The collar was designed to talk directly to the internet over 3G.

An API layer was used to create the data gateway that allowed the collar to connect with the back-end data collection and operational systems. At the time of writing, these collars were collecting 5 billion data points weekly so an ability to scale was very important.

Operational efficiency: benchmarking manufacturing

One of the world’s largest CPG companies has a seemingly simple problem. They have about two hundred factories, and each one has a set of machines that gathers information about the manufacturing process. This company had no easy way of getting that data from the factory locations to their head office. Using Anypoint Edge, the business could connect its sites and instantly collect data through an Edge Layer – using it for real-time reporting and benchmarking.   

APIs at the centre of IoT architecture

The ability to support devices beyond mobile applications within the enterprise is increasingly important. IoT will extend your existing value propositions and help drive new products and experiences. Devices will become woven into the fabric of your business connecting with operational systems to link the device data to CRM, ERP, or BI.

Having the composable building blocks to enable information flows from new sources is a necessity as is the ability to manage and govern those flows. You need composable building blocks to make this work. APIs are at the center of IoT architecture. The emergence of IoT has extended the notion that APIs can connect anything to everything.

The CIO’s role in developing an IoT strategy comes in imagining the successful outcome and building an IoT strategy, using an API-led approach to enable it.

Ross Mason, founder of MuleSoft

This is the third abridged extract from Ross Mason’s new book ‘First, Break IT’, available for free download.

Image source: Shutterstock/everything possible

Ross Mason
Ross Mason is MuleSoft’s founder. He leads engineering alignment, product strategy and direct engagement with customers at MuleSoft, provider of the leading platform for building application networks.