ophidiabigdata/ophidia-training repository overview

⁠All-in-one Ophidia HPDA Framework for training

This image contains a ready-to-use Ophidia HPDA framework⁠ instance that can be used for testing and training purposes. Besides the whole Ophidia software stack and related dependencies, it includes the Jupyter Notebook application and a set of scientific Python modules (e.g., NumPy, SciPy, matplotlib, Cartopy), which can be used to run data analytics examples with PyOphidia (the Ophidia Python bindings).

The source of this image can be found on the GitHub repository: https://github.com/OphidiaBigData/ophidia-container/tree/main/single-node⁠

⁠Get the image

This image can be downloaded from the Docker Hub Registry with:

$ docker pull ophidiabigdata/ophidia-training:latest

Alternately the image can be built from the source files:

$ docker build -t ophidiabigdata/ophidia-training:test \
--build-arg jupyter=yes \
'https://github.com/OphidiaBigData/ophidia-container.git#main:single-node'

⁠Start the container

To start the container the following command can be used:

$ docker run --rm -it -v HOST_PATH:/home/ophidia/CONTAINER_PATH -p 8888:8888 ophidiabigdata/ophidia-training:latest

The working dir for Jupyter Notebook is set to "/home/ophidia". It is hence recommended to mount the example notebooks and input files from the host under this folder to access them from the UI, with the option -v HOST_PATH:/home/ophidia/CONTAINER_PATH. "HOST_PATH" should be formatted based on the OS path format.

The internal port used by the Jupyter Notebook server can be mapped to the host one with the -p 8888:8888 option.

Environment variables can be optionally specified:

• UI_PORT: the port where the Jupyter Notebook will be started. Default port is 8888.
• MEMORY: the memory (RAM limit in MB) used to configure the Ophidia components. Default memory allocation is 2GB.
• DEBUG: a flag to enable the debug mode. Not active by default.

For example:

$ docker run --rm -it -v HOST_PATH:/home/ophidia/CONTAINER_PATH \
-e UI_PORT=8889 \
-e MEMORY=4096 \
-e DEBUG=1 \
-p 8888:8889 \
ophidiabigdata/ophidia-training:latest

The Jupyter Notebook server can be accessed from the browser and the URL will be http://CONTAINER_IP:UI_PORT (for example: http://172.17.0.2:8888/⁠). The URL will also be shown in the output of the running container. If the CONTAINER_IP is not reachable try with http://localhost:UI_PORT To login the Jupyter Notebook server type "ophidia" as password.

⁠Licensing information

The Ophidia HPDA Framework is released under the GPLv3 license. Besides Ophidia, the image contains additional software dependencies and components. A partial list of software contained in the image and the related license info can be found at: https://github.com/OphidiaBigData/ophidia-container/blob/main/single-node/NOTICE.md⁠

As with all Docker images, these likely also contain other software which may be under other licenses (such as Bash, etc from the base distribution, along with any direct or indirect dependencies of the primary software being contained).

It is the image user's responsibility to ensure that any use of this image complies with any relevant licenses for all software contained within.