This repo contains the code in the form of a notebook to benchmark kafka clients.
we will use conda and pip install your env.
conda create -n kafka-benchmark python=3 ipython jupyter -y
source activate kafka-benchmark
conda install -c activisiongamescience confluent-kafka pykafka -y # will also get librdkafka
pip install kafka-python # pure python version is easy to install
You will also need a running kafka cluster. Provided is a docker-compose file to run one locally but you can spin up anyway you want.