workshoppdsw252025

Quantifying AWS S3 I/O Performance Boundaries Using the Roofline Model

Meng Tang, Zhihao Zhu, Luanzheng Guo, James G. Bandy, Timothy Carlson, Sarah Neuwirth, Anthony Kougkas, Xian-He Sun, Nathan R. Tallent

The 10th International Parallel Data Systems Workshop · ACM

DOI ↗

Abstract

Cloud storage systems such as AWS S3 are increasingly used for scientific computing workloads. This paper applies the roofline model to analyze and quantify the I/O performance boundaries of AWS S3 and identifies performance optimization opportunities for scientific applications using cloud object storage.

Connected work

Iowarp

Citation

@inproceedings{tang2025awss3,
  author = {Tang, Meng and Zhu, Zhihao and Guo, Luanzheng and Bandy, James G. and Carlson, Timothy and Neuwirth, Sarah and Kougkas, Anthony and Sun, Xian-He and Tallent, Nathan R.},
  booktitle = {The 10th International Parallel Data Systems Workshop},
  title = {Quantifying AWS S3 I/O Performance Boundaries Using the Roofline Model},
  year = {2025},
  month = nov,
  publisher = {ACM},
  volume = {},
  number = {},
  pages = {},
  keywords = {I/O, Cloud Storage, AWS S3, Roofline Model, Performance Analysis},
  doi = {10.1145/3731599.3767513},
  url = {https://doi.org/10.1145/3731599.3767513},
}