Developers favoring AWS, Microsoft Azure for cloud IoT platforms

An Eclipse Foundation IoT developer survey reveals how AWS and Microsoft are widening an early IoT lead potentially at the expense of Google Cloud Platform.

Amazon Web Services and Microsoft Azure are widening their collective lead as the Internet of things platforms as Google Cloud Platform loses some momentum, according to a developer survey by the Eclipse Foundation.

The Eclipse Foundation's fourth annual IoT developer survey found that 51.8 percent of developers cited AWS as their IoT cloud platform followed by 31.2 percent for Microsoft Azure. Google Cloud Platform was 18.79 percent, down a bit from 2017. Developers often use more than one IoT cloud platform so the responses don't add up to 100 percent. Kubernetes came from nowhere in 2017 to top 12 percent in 2018.

Mike Milinkovich, executive director of the Eclipse Foundation, said: "there's clearly cloud growth for IoT and public cloud is the predominant scenario and you can see the growth in AWS and Azure."

Here's a look.

Another trend worth noting in that chart is that private cloud and on-premises IoT deployments are losing favor to public cloud platforms. Benjamin Cabé, program manager and evangelist at Eclipse IoT, said there are some nuances in the on-prem and public cloud breakdown. For instance, Kubernetes' surge in the developer community could have been on-prem or in the cloud. Meanwhile, OpenStack has fared well and could be deployed in cloud or on-prem.

Eclipse's cloud platform question revolved around whether enterprises are using or plan to use a particular cloud platform. Perhaps the most notable item in the following slide is how GE Predix has fallen out of the mix. In 2017, GE Predix was used in 5.17 percent of deployments for IoT. Today, GE Predix is at 2.1 percent.

Top cloud providers 2018: How AWS, Microsoft, Google, IBM, Oracle, Alibaba stack up

Not surprisingly, the Eclipse survey had an open source bent to it. The survey, which had 502 people participate, did reveal what IoT developers were hoping to do, what data was being collected and what the most popular tools of the moment were.

Here are some key takeaways.

• Time series data is the most collected data type followed by device information and log data.
• The top IoT concerns were security, data collection, and analytics, connectivity and hardware integration.
• IoT platforms are being built the most followed by home automation systems and industrial automation. Smart cities, energy management, and agriculture rounded out the top 6.
• Raspbian, Ubuntu, and Debian were the top IoT operating systems and Linux distros. Linux was used by 71.8 percent of respondents in IoT devices.
• Java is the programming language of choice in IoT. Milinkovich noted that IoT programming languages shouldn't be fancy so that means traditional methods in many cases. "There's no compelling rationale for a new programming language for IoT," he said. "When dealing with gateways and clouds you're looking into the skillsets of what developers have today. Inserting something new isn't going to help."

MySQL was the dominant IoT database.

SOURCE BY