What news from AWS re:Invent last week will have the most impact on you?
Amazon Q, an AI chatbot for explaining how AWS works.
Super-fast S3 Express storage.
New Graviton 4 processor instances.
Emily Freeman leaving AWS.
I don't use AWS, so none of this will affect me.
API Management / Edge Computing / Tech Life

The Different Flavors of IoT APIs

Sep 30th, 2014 10:50pm by
Featued image for: The Different Flavors of IoT APIs
Feature image via Flickr Creative Commons.

Internet of Things APIs are gaining a lot of attention lately and companies are jumping into the fray with both feet. Vendors offer a variety of methods and capabilities that allow devices to talk to each other and cloud/Web-based services. The picture is a bit blurry at the moment, at least in my mind, on how all these things will pan out or even how they work in the real World. You better start taking a look now, because it will only get more complex as time goes on. Here are a few of the companies and services they provide.


The recently rolled-out Zetta platform, from Apigee wants to be a kind of a mediator of APIs for all kinds of Internet of Things (IoT) devices. One of the big problems facing the IoT industry is that creating and maintaining APIs is quite demanding. Zetta has parts at the device and in the cloud that help manage that complexity.

Using a hub and spoke model Apigee consolidates communication from and between individual devices, connecting them to a cloud infrastructure through servers and bridges. The intelligence of interacting with all the individual device nodes is abstracted up into the cloud, so developers can focus on building the services and applications, without having to build and rebuild all the little APIs out on the devices or at the edge. Zetta also takes care of transforming the different protocols between the devices, servers, and cloud.

Naturally, setting up the cloud infrastructure to handle node abstraction costs money. Apigee plans to cover the back-end infrastructure by providing various paid online services to customers. The software runs on inexpensive hardware such as the Beaglebone and Raspberry Pi.


The company views the connected home as about the devices, hubs and network within it. But it also views it as including two other dimensions: “accessing and controlling the home from outside of it’s walls (inside-out) and also pulling data, media, services into homes (outside-in).”


Another API arbiter is called Temboo. This platform acts as a layer on top of third-party APIs, using code snippets to trigger complex processes that run through their cloud platform. Code snippets are added to your device code, perhaps on an Arduino Yun, and present a common methodology of function calls across a broad range of APIs.

Code snippets are same format between different APIs. Temboo also tries to shield developers from having to maintain APIs on each device. If you know how to use Temboo for one application, you know how to use it for all.


Xively uses free and open libraries to let you connect different types of hardware, using a variety of languages to their cloud service. They provide a standards-based API over HTTP, sockets, and MQTT to make it easy to connect to the Internet of Things.

My view is that Xively is a little closer to the hardware, than some of the other systems we’ve talked about. Check out their products and services on their web site.


This company offers a way to connect your devices to a cloud service, much like Xively, using open APIs, real-time data collection, device status messages, and so on.

TimeControl is an interesting service ThingSpeak offers. This service activates any type of HTTP call at a pre-defined time from the user’s account in the cloud. This opens up lots of possibilities to poll devices and gather data on a regular basis. Using custom HTTP headers, GETs, POSTs, PUTs, and DELETEs, the platform can trigger all kinds of responses and actions on remote individual hardware nodes.


The Crowsnest platform uses local servers to connect individual devices to the cloud. The company’s current level of development handles cameras on a Raspberry Pi. Other devices are sure to follow and they plan to use a plug-in model for supported devices. Once a plug-in for a device is created, it’s available to all Crowsnest developers, for use in their projects.

Where Do We Go From Here?

I still think the Internet of Things is fairly immature and unrefined. Vendors, both device manufacturers and those offering services are trying to figure how to make money in this game. The companies outlined in this story are a good start. And, we certainly have a bright future for opportunities as far as hardware goes. Now is the time to get in on the fun and start building the future, with the Internet of Things.

Group Created with Sketch.
THE NEW STACK UPDATE A newsletter digest of the week’s most important stories & analyses.