DeviceHive is an Open Source IoT Data Platform which helps to connect devices to the cloud in minutes allowing to stream device data and send commands. DeviceHive is highly scalable through containerization. You can run a DeviceHive stack with single instance of each component, then scale up by adding additional Frontend, Backend, Kafka and ZooKeeper instances. And finally attach Apache Spark analytics to Apache Kafka.
The easiest way to try DeviceHive locally or in your development datacenter is to deploy it using Docker Compose.
This will start complete DeviceHive service stack running:
More details in the rdbms-image subdirectory.
Installation was tested on machine with CentOS 7 distribution.
DeviceHive can be installed on Kubernetes with provided Helm chart. This chart also installs PostgreSQL chart and Kafka chart from Kubeapps repositories. External installations of PostgreSQL and Kafka are not supported at the moment.
Previous installation method on Kubernetes using a kubectl utility and a plain YAML files are deprecated now. Please issue a ticket in our GitHub repository if you have questions about mirgating such environment to the one deployed with Helm chart.
If you like to try DeviceHive using Docker for Windows or Docker for Mac, please note that this software runs Docker in special Virtual Machine (that got automaticaly created for you by installer). By default these Virtual Machines with much lower parameters that required for DeviceHive, 2GB of RAM and 2 vCPU. Here is example of how to change parameters in Docker for Windows, on Macs this should be similar: