ska_sdp_batchlet.app.cli module

The batchlet is a tool useful to run dask-distributed batch processing pipelines on Slurm HPC clusters.

Batchlet acts as a wrapper tool over the pipeline process, and provides following abilities:

1. Managing the dask cluster used by the pipeline to perform dask-based computations 2. Monitoring the hardware resources 3. Monitoring the pipeline logs and generating events for the consumers

ska_sdp_batchlet.app.cli.run_pipeline(config)

Run batchlet with given configuration.

Parameters:

config (str) -- Either a path to a JSON config file or literal "-". If the string is "-" the configuration is read from stdin.

Return type:

None

ska_sdp_batchlet.app.cli.register_sub_commands(parser)

Register entry points as sub-commands for a given argument parser.

Parameters:

parser (argparse._SubParsersAction) -- The sub-parser object to which new commands will be added.

Raises:

TypeError -- If a loaded entry point is not an instance of argparse.ArgumentParser.

ska_sdp_batchlet.app.cli.main()

Command line entrypoint for batchlet.

Return type:

None