If you find PMWeaver useful in your research, please cite:

Suyash Mahar, Sihang Liu, Korakit Seemakhupt, Vinson Young, Samira Khan
Write Prediction for Persistent Memory Systems
30th International Conference on Parallel Architectures and Compilation Techniques (PACT), 2021

BibTex

@inproceedings{mahar2021write,
  title={Write Prediction for Persistent Memory Systems},
  author={Mahar, Suyash and Liu, Sihang and Seemakhupt, Korakit and Young, Vinson and Khan, Samira},
  booktitle={2021 30th International Conference on Parallel Architectures and Compilation Techniques (PACT)},
  pages={242--257},
  year={2021},
  organization={IEEE}
}

Artifact repository for PMWeaver.

This readme contains basic commands to reproduce results of the experiments. For more details, check the appendix of the paper.

PMWeaver Docker image

1. Build the docker image

docker build -t pmweaver .

2. Start the docker container

docker run -i -t -p 8888:8888 pmweaver /bin/bash

Running experiments

To run partX:

scripts/run_partX.py

To check progress:

scripts/helper_scripts/check_progress.sh

To kill all stray processes between runs (doesn't delete data):

scripts/helper_scripts/kill_stray.sh

To plot results of partX:

scripts/plot_scripts/plot_partX.py

To start an HTTP server to view results (Located at: scripts/plots/ae/):

scripts/helper_scripts/start_server.sh