Getting Started

Prerequisites

pip install kiso
pip install kiso[chameleon]

Running the experiment

Create application credentials for CHI@Edge as edge-app-cred-oac-edge-openrc.sh and CHI@TACC as tacc-app-cred-oac-edge-openrc.sh and place the files in the secrets directory.

Running the experiment

# Check Kiso experiment configuration
kiso check

# Provision and setup the resources
kiso up

# Run the experiments defined in the experiment configuration YAML file
kiso run

# Destroy the provisioned resources
kiso down

# Pegasus workflow submit directories will be placed in the output directory at the end of the experiment. The submit directories will also have a statistics directory with the pegasus-statistics output.
# Outputs defined in the experiment configuration will be placed in the destination specified in the experiment configuration.

References

• Pegasus Workflow Management System
• EnOSlib
• Chameleon Cloud
• FABRIC

Citation

• Kyathanahally, S. P., Hardeman, T., Merz, E., Bulas, T., Reyes, M., Isles, P., Pomati, F., & Baity-Jesi, M. (2021). Deep Learning Classification of Lake Zooplankton. Frontiers in Microbiology, 12.

Acknowledgements

Kiso is funded by National Science Foundation (NSF) under award 2403051.