Skip to content

Know Your Enemy To Save Cloud Energy: Energy-Performance Characterization of Machine Learning Serving (HPCA '23)

License

Notifications You must be signed in to change notification settings

EMDC-OS/power-aware-triton

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Know Your Enemy To Save Cloud Energy:
Energy-Performance Characterization of Machine Learning Serving

This is the artifact repository for the HPCA'23 paper "Know Your Enemy To Save Cloud Energy: Energy-Performance Characterization of Machine Learning Serving". The prototype and the simulation code are available.

Contents

Contents

Prototype Inference Server

The prototype inference server is based on the Triton Inference Server v2.7.0 (TIS). The source code is available here.

Cloud-scale Simulation

The simulation was performed based on the values obtained from the prototype to observe the effectiveness of the proposed schemes in the cloud-scale environment. The source code is available here.

Citation

@INPROCEEDINGS{yu2023know,
  author={Yu, Junyeol and Kim, Jongseok and Seo, Euiseong},
  title={Know Your Enemy To Save Cloud Energy: Energy-Performance Characterization of Machine Learning Serving},
  booktitle={2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)},
  year={2023},
  pages={842-854},
  doi={10.1109/HPCA56546.2023.10070943}}

Contact

If you have any questions, please contact [email protected]

About

Know Your Enemy To Save Cloud Energy: Energy-Performance Characterization of Machine Learning Serving (HPCA '23)

Resources

License

Stars

Watchers

Forks