k-python-boinc

Container for running your boinc in KnowledgeX

Non TEE workflow

Paste your code into the def your_python_code() in the template_app.py.

Run locally with env variables

  IEXEC_INPUT_FILES_NUMBER=1 \
  IEXEC_IN=sample_data \
  IEXEC_INPUT_FILES_FOLDER=sample_data \
  IEXEC_INPUT_FILE_NAME_1=data_set_full.csv \
  IEXEC_OUT=output \
  python3 src/template_app.py

Build Docker image

docker build . --tag k-python-boinc

Run from locally build docker image

rm -rf output && \
docker run \
        --rm \
        -e IEXEC_IN=/iexec_in \
        -e IEXEC_OUT=/iexec_out \
        -e IEXEC_INPUT_FILES_NUMBER=0 \
        -e IEXEC_INPUT_FILES_FOLDER=/iexec_in \
        -v $(pwd)/output:/iexec_out \
        -v $(pwd)/sample_data:/iexec_in \
        k-python-boinc

Once the execution ends, the result should be found in the folder output.

cat output/computed.json

Deploy to DockerHub

docker tag k-python-boinc <dockerhub-user>/k-python-boinc:1.0.0
docker push <dockerhub-user>/k-python-boinc:1.0.0

Run docker image directly from dockerhub

rm -rf output
docker run \
        --rm \
        -e IEXEC_IN=/iexec_in \
        -e IEXEC_OUT=/iexec_out \
        -e IEXEC_INPUT_FILES_NUMBER=1 \
        -e IEXEC_INPUT_FILES_FOLDER=/iexec_in \
        -e IEXEC_INPUT_FILE_NAME_1=data_set_full.csv \
        -v $(pwd)/output:/iexec_out \
        -v $(pwd)/sample_data:/iexec_in \
        docker.io/jadenx/k-python-boinc:1.0.0