F-DATA

F-DATA: A Fugaku Workload Dataset for Job-centric Predictive Modelling in HPC Systems

This repository contains the scripts and documentation for the F-DATA, available in Zenodo .

Instruction on how to load the data

The files of F-DATA are saved as .parquet files. It is possible to load such files as dataframes by leveraging the pandas APIs, after installing pyarrow (pip install pyarrow). A single file can be loaded as follows:

# Importing pandas library
import pandas as pd 

# Read the 21_01.parquet file in a dataframe format
df = pd.read_parquet("21_01.parquet")
df.head()

Repository structure

• baseline_experiments.py: The script to execute ML predictive modelling on the F-DATA.
• generate_plots.py: The script to generate a series of plots.
• requirements.txt: The python dependencies to execute all the scripts in the repository.
• docs: The folder contains some documentation of the final dataset, such as the job feature list and description.
• plots : The folder contains the plots of the whole F-DATA, as well as of the single splits that can be found in Zenodo.
• generation_scripts: The folder contains the scripts used to anonymize the data and generate the derived features.

Contact us

For any information on F-DATA don't hesitate to contact us at: francesco.antici98[at]gmail.com.

Cite us

Please cite the work as

@article{antici2025fdata,
  title={F-DATA: A Fugaku Workload Dataset for Job-centric Predictive Modelling in HPC Systems},
  author={Antici, Francesco and Bartolini, Andrea and Domke, Jens and Kiziltan, Zeynep and Yamamoto, Keiji},
  journal = {Scientific Data},
  volume={12},
  pages={1321},
  year={2025},
  publisher={Nature Publishing Group},
  doi={https://doi.org/10.1038/s41597-025-05633-1}
}