Cloud computing

To get data together from several Astroplants throughout the world and to analyse these, we use a central Cloud platform at SURF. Here all data is stored and ready to visualise or analyse. Much information about this platform can be found at the Sensemakers Github under sensemakers-iot-platform.

Step 1: Sending the data

To send the data we use a protocol called MQTT. MQTT or Message Queue Telemetry Transport is a publish-subscribe protocol widely used in the Internet of Things. With this protocol we can send the sensor data to the SURF cloud platform.

The specific MQTT python module we use is paho-mqtt which you can install with 'pip3 install paho-mqtt'. The example script 'read_and_send_all.py' shows how to set up a connection and how to send data. Note that you have to specify a mqtt host, user and password in a separate file (credentials.py).

To make sure your data will be labeled as yours: you need to specify it as part of the 'topic' and as 'device_id'. In case of the example it is 'YourNameHere', which you of course have to replace with your own name. (note it occurs twice in the code).

In the back-end the data will be imported automatically into an Influx database

(Detailed documentation about paho mqtt can be found on https://www.eclipse.org/paho/clients/python/ )

Step 2: Visualising the data

When your data is in the database, you should be able to see it on grafana.sensemakersams.org. The data is public, but you can play with the devices to show and the time period of which you want to see data.

If you want to build your own dashboard (for instance to define gauges and graphs for other sensors you added ask David Salek, for a username and password for this.

Step 3: Analysing the data

Another way to use your sensor data is to access it with Jupyter notebooks and do calculations. The website is jupyter.sensemakersams.org. Again here you will need credentials to get access from David. A good way to start learning about Jupyter Notebooks is to do the courses on Data Visualisation and Pandas at Kaggle