Blog Post

Here’s a startup that wants to be an OpenStack for the internet of things

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

The co-founders of OpenSensors believe that while connecting billions of devices to the internet will be complicated, being able to do it on infrastructure you control should be easy. That’s why the London-based startup formed last year with plan to build a software tool that lets companies build out the infrastructure needed to host millions of devices.

For example, a city looking to put sensors in parking spaces and sensors on streetlights might have two different vendors contracted for the two different services, and may have different access to the resulting data. Thus, building an app that combined both elements or opening the data up to outside developers might not be possible. But if a city could buy sensors, install them and have a relatively easy way to get that data up in the cloud and then publish it, maybe it wouldn’t need to farm out the work — and control of the data — to a third party.

That’s what OpenSensors wants to help with. Yodit Stanton, a co-founder spoke with me last week to tell me about Azondi, a software that contains a database, some basic authentication and a publishing and subscription model using MQTT that can act as the beginnings of a scalable infrastructure on which to host sensors.

The idea is that a customer licenses the software and then can run his sensor network on his own servers, in a cloud, or where ever he’d like. The point is that the customer is in control. The Azondi software is available on GitHub and eventually Stanton anticipates people taking a license to it and forking the code for their own needs.

Stanton said that many customers looking at building out some type of sensor network find that the available cloud platforms such as Xively or Axeda, or even certain cloud platform provided by their hardware provider, aren’t transferable. So their data goes in, but only comes out in the form of an API at best. In some cases, getting data out via an API is a best case scenario. Stanton said she once went searching halfway around the world and through seven different vendors to get some parking data for the City of Westminster in London.

Both Stanton and her co-founder, Malcolm Sparks, are middleware experts who look at the internet of things and see a vast opportunity — not just for systems integration — but for a variety of sources of information to come together to make entirely new applications and services. They built Azondi to let municipalities and any other customers take as much control over that process as they want to.

4 Responses to “Here’s a startup that wants to be an OpenStack for the internet of things”


    I propose to you a variance on the scenario proposed above.

    -a company that is in the business of deploying endpoints, deploys the sensors.
    -That company subscribes to a billing platform to collect revenue from any one that gets access to the data from those sensors.
    -The city and any other interested party subscribes to the data via the billing platform.

    Why should the city get into the sensor deployment business?
    Why should the city carry the burden of the full cost of deployment?

    With the billing platform, any one can subscribe for daily, weekly or annual access. The cost is then spread across many users. Once the route to revenue is evident, many people would create endpoint companies and those companies will for sure offer the correct data format and APIs since their business depends on it.

    What do you think?

    • shparekh

      I agree cities, municipalities, businesses should not be investing in tech that has yet to see its major evolutionary phase. I think a utility that procures data from the sensor companies (end-points) and resells it to anyone that needs that data can succeed at this early stage of IoT.