In this blog post series, we will demonstrate how to use AWS IoT Greengrass and Hashicorp Nomad to seamlessly interface with multiple interconnected devices and orchestrate service deployments on them. Greengrass will allow us to view the cluster as a single "Thing" from the cloud perspective, while Nomad will serve as the primary cluster orchestration tool. - Source: dev.to / over 1 year ago

AWS IoT Greengrass allows one to manage their IOT Edge devices, download ML models locally, so that inference can then be also be done locally. - Source: dev.to / about 2 years ago

To assist in deployment and management of workloads in your fleet, it's worth taking advantage of a fleet or device management tool such as AWS GreenGrass, Formant or Rocos. - Source: dev.to / over 2 years ago

In some cases, such as with edge devices, inferencing needs to occur even when there is limited or no connectivity to the cloud. Mining fields are an example of this type of use case. AWS IoT Greengrass enables ML inference locally using models that are created, trained, and optimized in the cloud using Amazon SageMaker, AWS Deep Learning AMI, or AWS Deep Learning Containers, and deployed on the edge devices. - Source: dev.to / over 2 years ago

Take a look at Greengrass https://aws.amazon.com/greengrass/ Enables OTA updates and fleet management. Source: about 3 years ago

I can recommend [Zephyr](https://github.com/zephyrproject-rtos/zephyr) for these things. Source: over 2 years ago

Zephyr, RIOT, NuttX are all open source RTOSes happy about contributions. Source: over 2 years ago

Microcontrollers are a good start. Boards are cheap and sensors are plentiful. You could get a yet unsupported board and port RIOT, Zephyr or NuttX. Source: over 2 years ago

Zephyr Project: Update to 2.5 or later.  Patches are available for prior supported versions. See the Zephyr security advisory for more information. Source: almost 3 years ago