Geoffrey Yu

Geoffrey Yu

Computer Science PhD Student

Massachusetts Institute of Technology (MIT)

About Me

I am a Computer Science PhD student at MIT, advised by Professor Tim Kraska. I am part of the Data Systems Group within the Computer Science and Artificial Intelligence Laboratory (CSAIL).

I am generally interested in computer systems research. My research interests span data systems, distributed systems, and systems for machine learning. I also enjoy thinking about problems at the intersection of systems and human-computer interaction as I strongly believe in the value of creating usable systems software.

Before starting my PhD, I earned my master's degree in Computer Science at the University of Toronto, advised by Professor Gennady Pekhimenko. Before graduate school, I was a Software Engineering student at the University of Waterloo and graduated in 2018.

News

August 28, 2023
At VLDB 2023 in Vancouver, I presented our group's vision for BRAD and jointly presented TreeLine.
July 1, 2023
Our vision for a new unified cloud data processing system called BRAD was accepted to appear at VLDB 2023!
August 16, 2022
Our TreeLine paper will appear at VLDB 2023!
See older news »

Publications

Check Out the Big Brain on BRAD: Simplifying Cloud Data Processing with Learned Automated Data Meshes

Check Out the Big Brain on BRAD: Simplifying Cloud Data Processing with Learned Automated Data Meshes

Tim Kraska*, Tianyu Li*, Samuel Madden*, Markos Markakis*, Amadou Ngom*, Ziniu Wu*, Geoffrey X. Yu*.

Proceedings of the VLDB Endowment (VLDB), Vol. 16, No. 11., 2023. Vision Paper.

Paper | BibTeX | Project Website 

@article{brad-kraska23,
  author = {Kraska, Tim and Li, Tianyu and Madden, Samuel and Markakis, Markos
    and Ngom, Amadou and Wu, Ziniu and Yu, Geoffrey X.},
  doi = {10.14778/3611479.3611526},
  journal = {{Proceedings of the VLDB Endowment}},
  month = {8},
  number = {11},
  pages = {3293--3301},
  title = {{Check Out the Big Brain on BRAD: Simplifying Cloud Data Processing
    with Learned Automated Data Meshes}},
  volume = {16},
  year = {2023},
}
TreeLine: An Update-In-Place Key-Value Store for Modern Storage

TreeLine: An Update-In-Place Key-Value Store for Modern Storage

Geoffrey X. Yu*, Markos Markakis*, Andreas Kipf*, Per-Åke Larson, Umar Farooq Minhas, Tim Kraska.

Proceedings of the VLDB Endowment (VLDB), Vol. 16, No. 1., 2022.

Paper | BibTeX 

@article{treeline-yu23,
  author = {Yu, Geoffrey X. and Markakis, Markos and Kipf, Andreas and
    Larson, Per-Åke and Minhas, Umar Farooq and Kraska, Tim},
  doi = {10.14778/3561261.3561270},
  journal = {{Proceedings of the VLDB Endowment}},
  month = {9},
  number = {1},
  pages = {99--112},
  title = {{TreeLine: An Update-In-Place Key-Value Store for Modern Storage}},
  volume = {16},
  year = {2022},
}
Habitat: A Runtime-Based Computational Performance Predictor for Deep Neural Network Training

Habitat: A Runtime-Based Computational Performance Predictor for Deep Neural Network Training

Geoffrey X. Yu, Yubo Gao, Pavel Golikov, Gennady Pekhimenko.

USENIX Annual Technical Conference (USENIX ATC), 2021.

Paper | BibTeX | Talk Video | Project Website 

@inproceedings{habitat-yu21,
  title = {{Habitat: A Runtime-Based Computational Performance Predictor
    for Deep Neural Network Training}},
  author = {Yu, Geoffrey X. and Gao, Yubo and Golikov, Pavel and
    Pekhimenko, Gennady},
  booktitle = {{Proceedings of the 2021 USENIX Annual Technical Conference
    (USENIX ATC'21)}},
  year = {2021},
}
Skyline: Interactive In-Editor Computational Performance Profiling for Deep Neural Network Training

Skyline: Interactive In-Editor Computational Performance Profiling for Deep Neural Network Training

Geoffrey X. Yu, Tovi Grossman, Gennady Pekhimenko.

ACM Symposium on User Interface Software and Technology (UIST), 2020.

Paper | BibTeX | Project Website 

@inproceedings{skyline-yu20,
  title = {{Skyline: Interactive In-Editor Computational Performance Profiling
    for Deep Neural Network Training}},
  author = {Yu, Geoffrey X. and Grossman, Tovi and Pekhimenko, Gennady},
  booktitle = {{Proceedings of the 33rd ACM Symposium on User Interface
    Software and Technology (UIST'20)}},
  year = {2020},
}

* Denotes equal contribution.

Demonstrations

Skyline: Interactive In-Editor Performance Visualizations and Debugging for DNN Training

Geoffrey X. Yu, Tovi Grossman, Gennady Pekhimenko.

Machine Learning and Systems (MLSys), 2020. Demonstration Track, Non-archival.

TBD Suite: Benchmarking and Profiling Tools for DNNs

Geoffrey X. Yu, Hongyu Zhu, Anand Jayarajan, Bojian Zheng, Abhishek Tiwari, Gennady Pekhimenko.

Machine Learning and Systems (MLSys), 2019. Demonstration Track, Non-archival.

[PDF]