Skip to content

Latest commit

 

History

History
57 lines (46 loc) · 3.43 KB

README.md

File metadata and controls

57 lines (46 loc) · 3.43 KB
Raw

Apache StreamPipes for Pythonistas

This repository contains additional material for the talk Apache StreamPipes for Pythonistas: IIoT data handling made easy! given by @bossenti and @SvenO3 at PyConDE 2023, Berlin. It contains the following elements:

  • Presentation slides
  • Material for the first demo
  • Material for the second demo

First Demo - Ingest data (Modbus source)

The first demo shows how to use Apache StreamPipes to ingest data from an IIoT data source, in this case a Modbus device. This requires a running StreamPipes instance. There are three ways to start/deploy StreamPipes (CLI, Docker Compose, and Kubernetes), of which we recommend using Docker Compose for end users. All you need to do is run the following command in installer/compose within the StreamPipes repository/source code:

docker-compose -f docker-compose.nats.yaml up -d

For more detailed information about how to start StreamPipes, visit our docs.

The second precondition of this demo is a running Modbus device, which we simulate by using two Python scripts:

  • demo-one/modbus_server.py: This script runs a Modbus server capable of receiving data from producers and serving as data source
  • demo-one/modbus_data_insert.py: This script reads data from a given CSV file and publishes it to the given Modbus device.

To run both scripts, you need to ensure that the required dependencies are in place, as specified by demo-one/requirements.txt.. Then you can initialize the Modbus server via executing the following command:

python demo-one/modbus_server.py

The server runs until you terminate the Python process.

In the next step we want to read the data from the file demo-one/heat_pump.csv into the Modbus server with the following command:

python demo-one/modbus_data_insert.py --drop timestamp --file ./data/heat_pump_modbus.csv

The script will publish one row of the specified CSV file per second, drop all columns specified by the --drop' flag using 0' as the node identifier, and terminate when all rows are published.

The data provided in the data directory is real data from a heat pump (kudos to @SvenO3 for recording). Within the this directory, there are two files:

  • ..._modbus.csv: all metrics are multiplied by 100 since modbus can only handle integer values natively (needs to be taken in account when working with the data)
  • ..._float.csv: the very same data but already divided by 100 and therefore in the correct form

Now you can read the data from StreamPipes via a connection to the Modbus server listening on localhost:5002. Be aware that you need to pass host.docker.internal as host inside StreamPipes.

Second Demo - StreamPipes Python

This demo is all about demonstrating how to use the StreamPipes Python library. All one need to is, to follow the Jupyter notebook in demo-two. There you can see how to setup the StreamPipes client to interact with the API and to implement a simple StreamPipes Function that allows you to interact with live data.

Talk Video

Please find the talk's recording on YouTube:

Youtube Video - Apache StreamPipes for Pythonistas: IIoT data handling made easy!.