Three Things I Hope to Learn From the 2021 Eclipse IoT and Edge Developer Survey

The 2021 edition of the Eclipse IoT & Edge Developer Survey is underway. This is the seventh edition of the survey, which over the years has become one of the most widely referenced technical surveys within the IoT & edge computing community.

Back in July, strategic members of the Eclipse IoT, Edge Native and Sparkplug working groups were overhauling the survey. As a result, I sincerely think this is our best survey yet, and I am excited about our upcoming report on the results, which we intend to publish in November.

As usual, the survey results will be chock-full of insights. Here are three questions to which I look forward to seeing answers.

How are developers leveraging edge gateways and edge servers?

In past editions of the survey, we have asked developers which languages they use and which workloads they deploy on edge nodes. This year, we made this question more granular as we split edge nodes into edge gateways and edge servers. What is the distinction, you ask? An edge gateway acts as the aggregation point for a group of sensors and actuators to coordinate the connectivity of these devices to each other and an external network. On the other hand, an edge server can run multiple workloads and is, by design, meant to offer workload consolidation at the edge.

Given their distinct purposes, I would expect some differences in the way developers configure edge gateways and edge servers. I hope the survey results will shed some light on this question.

Sparkplug is on the rise, but by how much?

Year after year, our respondents have selected MQTT as the number one option among protocols specific to IoT. This is no surprise. MQTT is mature and proven. The Eclipse Paho and Eclipse Mosquitto projects are fantastic client and broker implementations, respectively. And with Eclipse Amlen, there will now be an open-source option for those looking for a highly available broker that supports clustering.

What makes MQTT great is that it is very flexible since the specification does not define the message payloads. However, this flexibility comes at the price of interoperability. Fortunately, there is Eclipse  Sparkplug, a protocol that leverages MQTT as its transport. By defining standard payloads and topic structures, Sparkplug makes MQTT-based infrastructure interoperable out of the box. 

The 2020 edition of the survey found significant adoption for Sparkplug. One year later, I am pretty sure it has seen substantial growth. The only question is: by how much?

How are CORE-V and RISC-V impacting the industry?

Even Apple recently posted jobs related to RISC-V. So I am confident we'll see some growth for the open source ISA this year. However, given the current market share of ARM-based SOCs and MCUs, I am unsure how RISC-V's current momentum will translate into our IoT and Edge community.

We added the CORE-V family of RISC-V cores to our questions about CPUs for constrained devices and edge gateways/servers for the first time. Our friends at OpenHW Group have been busy working on their open source cores, and it will be interesting to explore its impact on the developer community further.

Make your voice heard

The 2021 edition of the Eclipse IoT and Edge Developer Survey will be open until October 5, 2021. As usual, we will publish our findings under a Creative Commons license to benefit the open source community at large. Please complete the survey now and make your voice heard!