Publications | Jie Zhao

2024

ISSTA 2024

Arfa: An Agile Regime-Based Floating-Point Optimization Approach for Rounding Errors

Jinchen Xu, Mengqi Cui, Fei Li, Zuoyan Zhang, Hongru Yang, Bei Zhou, and Jie Zhao

In Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis, Vienna, Austria, 2024

Bib PDF Slides

@inproceedings{arfa,
  author = {Xu, Jinchen and Cui, Mengqi and Li, Fei and Zhang, Zuoyan and Yang, Hongru and Zhou, Bei and Zhao, Jie},
  title = {Arfa: An Agile Regime-Based Floating-Point Optimization Approach for Rounding Errors},
  year = {2024},
  isbn = {9798400706127},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3650212.3680378},
  doi = {10.1145/3650212.3680378},
  booktitle = {Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis},
  pages = {1516--1528},
  numpages = {13},
  keywords = {FPbench, dynamic analysis, e-graph, floating-point errors, numerical analysis, rewrite},
  location = {Vienna, Austria},
  series = {ISSTA 2024}
}

OSDI 2024

Enabling Tensor Language Model to Assist in Generating High-Performance Tensor Programs for Deep Learning

Yi Zhai, Sijia Yang, Keyu Pan, Renwei Zhang, Shuo Liu, Chao Liu, Zichun Ye, Jianmin Ji, Jie Zhao, Yu Zhang, and Yanyong Zhang

In 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24), Jul 2024

Bib PDF Slides

@inproceedings{tlm,
  author = {Zhai, Yi and Yang, Sijia and Pan, Keyu and Zhang, Renwei and Liu, Shuo and Liu, Chao and Ye, Zichun and Ji, Jianmin and Zhao, Jie and Zhang, Yu and Zhang, Yanyong},
  title = {Enabling Tensor Language Model to Assist in Generating {High-Performance} Tensor Programs for Deep Learning},
  booktitle = {18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24)},
  year = {2024},
  isbn = {978-1-939133-40-3},
  address = {Santa Clara, CA},
  pages = {289--305},
  url = {https://www.usenix.org/conference/osdi24/presentation/zhai},
  publisher = {USENIX Association},
  month = jul
}

PPoPP 2024

A Holistic Approach to Automatic Mixed-Precision Code Generation and Tuning for Affine Programs

Jinchen Xu, Guanghui Song, Bei Zhou, Fei Li, Jiangwei Hao, and Jie Zhao

In Proceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, , Edinburgh, United Kingdom, , Jul 2024

Bib PDF Slides

@inproceedings{prectuner,
  author = {Xu, Jinchen and Song, Guanghui and Zhou, Bei and Li, Fei and Hao, Jiangwei and Zhao, Jie},
  title = {A Holistic Approach to Automatic Mixed-Precision Code Generation and Tuning for Affine Programs},
  year = {2024},
  isbn = {9798400704352},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3627535.3638484},
  doi = {10.1145/3627535.3638484},
  booktitle = {Proceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming},
  pages = {55–67},
  numpages = {13},
  keywords = {polyhedral model, performance, autotuning},
  location = {<conf-loc>, <city>Edinburgh</city>, <country>United Kingdom</country>, </conf-loc>},
  series = {PPoPP '24}
}

TOCS 2024

Modeling the Interplay between Loop Tiling and Fusion in Optimizing Compilers Using Affine Relations

Jie Zhao, Jinchen Xu, Peng Di, Wang Nie, Jiahui Hu, Yanzhi Yi, Sijia Yang, Zhen Geng, Renwei Zhang, Bojie Li, Zhiliang Gan, and Xuefeng Jin

ACM Trans. Comput. Syst., Jan 2024

Bib PDF

@article{basteln,
  author = {Zhao, Jie and Xu, Jinchen and Di, Peng and Nie, Wang and Hu, Jiahui and Yi, Yanzhi and Yang, Sijia and Geng, Zhen and Zhang, Renwei and Li, Bojie and Gan, Zhiliang and Jin, Xuefeng},
  title = {Modeling the Interplay between Loop Tiling and Fusion in Optimizing Compilers Using Affine Relations},
  year = {2024},
  issue_date = {November 2023},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {41},
  number = {1--4},
  issn = {0734-2071},
  url = {https://doi.org/10.1145/3635305},
  doi = {10.1145/3635305},
  journal = {ACM Trans. Comput. Syst.},
  month = jan,
  articleno = {5},
  numpages = {45},
  keywords = {optimizing compilers, polyhedral model, memory hierarchy, redundant computation, parallelism, data locality, fusion, Tiling}
}

2023

MLSys 2023

SIRIUS: Harvesting Whole-Program Optimization Opportunities for DNNs

Yijin Li, Jiacheng Zhao, Qianqi Sun, Haohui Mai, Lei Chen, Wanlu Cao, Yanfan Chen, Zhicheng Li, Ying Liu, Xinyuan Zhang, Xiyu Shi, Jie Zhao, Jingling Xue, Huimin Cui, and Xiaobing Feng

In Proceedings of Machine Learning and Systems, Jan 2023

Bib PDF Slides

@inproceedings{sirius,
  author = {Li, Yijin and Zhao, Jiacheng and Sun, Qianqi and Mai, Haohui and Chen, Lei and Cao, Wanlu and Chen, Yanfan and Li, Zhicheng and Liu, Ying and Zhang, Xinyuan and Shi, Xiyu and Zhao, Jie and Xue, Jingling and Cui, Huimin and Feng, Xiaobing},
  booktitle = {Proceedings of Machine Learning and Systems},
  editor = {Song, Dawn and Carbin, Michael and Chen, Tianqi},
  pages = {1-19},
  title = {SIRIUS: Harvesting Whole-Program Optimization Opportunities for DNNs},
  volume = {5},
  year = {2023}
}

OSDI 2023

Effectively Scheduling Computational Graphs of Deep Neural Networks toward Their Domain-Specific Accelerators

Jie Zhao, Siyuan Feng, Xiaoqiang Dan, Fei Liu, Chengke Wang, Sheng Yuan, Wenyuan Lv, and Qikai Xie

In Proceedings of the 17th USENIX Symposium on Operating Systems Design and Implementation (OSDI 23), Jul 2023

Bib PDF Slides

@inproceedings{graphturbo,
  author = {Zhao, Jie and Feng, Siyuan and Dan, Xiaoqiang and Liu, Fei and Wang, Chengke and Yuan, Sheng and Lv, Wenyuan and Xie, Qikai},
  title = {Effectively Scheduling Computational Graphs of Deep Neural Networks toward Their Domain-Specific Accelerators},
  booktitle = {Proceedings of the 17th USENIX Symposium on Operating Systems Design and Implementation (OSDI 23)},
  year = {2023},
  isbn = {978-1-939133-34-2},
  address = {Boston, MA},
  pages = {719--737},
  url = {https://www.usenix.org/conference/osdi23/presentation/zhao},
  publisher = {USENIX Association},
  month = jul,
}

ASE 2023

Eiffel: Inferring Input Ranges of Significant Floating-point Errors via Polynomial Extrapolation

Zuoyan Zhang, Bei Zhou, Jiangwei Hao, Hongru Yang, Mengqi Cui, Yuchang Zhou, Guanghui Song, Fei Li, Jinchen Xu, and Jie Zhao

In Proceedings of the 38th IEEE/ACM International Conference on Automated Software Engineering (ASE 2023), Sep 2023

Bib PDF Slides

@inproceedings{eiffel,
  author = {Zhang, Zuoyan and Zhou, Bei and Hao, Jiangwei and Yang, Hongru and Cui, Mengqi and Zhou, Yuchang and Song, Guanghui and Li, Fei and Xu, Jinchen and Zhao, Jie},
  title = {Eiffel: Inferring Input Ranges of Significant Floating-point Errors via Polynomial Extrapolation},
  booktitle = {Proceedings of the 38th IEEE/ACM International Conference on Automated Software Engineering (ASE 2023)},
  year = {2023},
  isbn = {978-1-939133-34-2},
  address = {Kirchberg, Luxembourg},
  pages = {1441--1453},
  publisher = {IEEE Press},
  month = sep,
}

2022

PACT 2022

Parallelizing Neural Network Models Effectively on GPU by Implementing Reductions Atomically

Jie Zhao, Cédric Bastoul, Yanzhi Yi, Jiahui Hu, Wang Nie, Renwei Zhang, Zhen Geng, Chong Li, Thibaut Tachon, and Zhiliang Gan

In Proceedings of 31st International Conference on Parallel Architectures and Compilation Techniques, Chicago, IL, USA, Sep 2022

Bib PDF Slides

@inproceedings{panamera,
  location = {Chicago, IL, USA},
  address = {New York, NY, USA},
  author = {Zhao, Jie and Bastoul, C\'{e}dric and Yi, Yanzhi and Hu, Jiahui and Nie, Wang and Zhang, Renwei and Geng, Zhen and Li, Chong and Tachon, Thibaut and Gan, Zhiliang},
  booktitle = {Proceedings of 31st International Conference on Parallel Architectures
  and Compilation Techniques},
  isbn = {9781450398688},
  pages = {451-466},
  publisher = {Association for Computing Machinery},
  series = {PACT'22},
  title = {Parallelizing Neural Network Models Effectively on GPU by Implementing Reductions Atomically},
  doi = {https://doi.org/10.1145/3559009.3569656},
  year = {2022}
}

ICPP 2022

Automatically Generating High-performance Matrix Multiplication Kernels on the Latest Sunway Processor

Xiaohan Tao, Yu Zhu, Boyang Wang, Jinlong Xu, Jianmin Pang, and Jie Zhao

In Proceedings of the 51st International Conference on Parallel Processing, Bordeaux, France, Sep 2022

Bib PDF Slides

@inproceedings{swgemm,
  location = {Bordeaux, France},
  address = {New York, NY, USA},
  author = {Tao, Xiaohan and Zhu, Yu and Wang, Boyang and Xu, Jinlong and Pang, Jianmin and Zhao, Jie},
  booktitle = {Proceedings of the 51st International Conference on Parallel Processing},
  isbn = {9781450397339},
  numpages = {12},
  pages = {23:1-23:12},
  publisher = {Association for Computing Machinery},
  series = {ICPP'22},
  title = {Automatically Generating High-performance Matrix Multiplication Kernels on the Latest Sunway Processor},
  doi = {https://doi.org/10.1145/3545008.3545031},
  year = {2022}
}

MLSys 2022

Apollo: Automatic Partition-based Operator Fusion through Layer by Layer Optimization

Jie Zhao, Xiong Gao, Ruijie Xia, Zhaochuang Zhang, Deshi Chen, Lei Chen, Renwei Zhang, Zhen Geng, Bin Cheng, and Xuefeng Jin

In Proceedings of Machine Learning and Systems, Sep 2022

Bib PDF Slides

@inproceedings{apollo,
  author = {Zhao, Jie and Gao, Xiong and Xia, Ruijie and Zhang, Zhaochuang and Chen, Deshi and Chen, Lei and Zhang, Renwei and Geng, Zhen and Cheng, Bin and Jin, Xuefeng},
  booktitle = {Proceedings of Machine Learning and Systems},
  date-modified = {2022-02-19 09:27:32 +0800},
  editor = {Marculescu, Diana and Chi, Yuejie and Wu, Carole-Jean},
  pages = {1-19},
  title = {Apollo: Automatic Partition-based Operator Fusion through Layer by Layer Optimization},
  volume = {4},
  year = {2022}
}

2021

PLDI 2021

AKG: Automatic Kernel Generation for Neural Processing Units Using Polyhedral Transformations

Jie Zhao, Bojie Li, Wang Nie, Zhen Geng, Renwei Zhang, Xiong Gao, Bin Cheng, Chen Wu, Yun Cheng, Zheng Li, Peng Di, Kun Zhang, and Xuefeng Jin

In Proceedings of the 42nd ACM SIGPLAN International Conference on Programming Language Design and Implementation, Sep 2021

Bib PDF Slides

@inproceedings{akg,
  address = {New York, NY, USA},
  author = {Zhao, Jie and Li, Bojie and Nie, Wang and Geng, Zhen and Zhang, Renwei and Gao, Xiong and Cheng, Bin and Wu, Chen and Cheng, Yun and Li, Zheng and Di, Peng and Zhang, Kun and Jin, Xuefeng},
  booktitle = {Proceedings of the 42nd ACM SIGPLAN International Conference on Programming Language Design and Implementation},
  isbn = {9781450383912},
  numpages = {16},
  pages = {1233--1248},
  publisher = {Association for Computing Machinery},
  series = {PLDI'21},
  title = {AKG: Automatic Kernel Generation for Neural Processing Units Using Polyhedral Transformations},
  url = {https://doi.org/10.1145/3453483.3454106},
  year = {2021},
  bdsk-url-1 = {https://doi.org/10.1145/3453483.3454106}
}

2020

MICRO 2020

Optimizing the Memory Hierarchy by Compositing Automatic Transformations on Computations and Data

Jie Zhao, and Peng Di

In Proceedings of the 53rd IEEE/ACM International Symposium on Microarchitecture, Virtual Event, Athens, Greece, Sep 2020

Bib PDF Slides

@inproceedings{polytef,
  address = {Piscataway, NJ, USA},
  author = {Zhao, Jie and Di, Peng},
  booktitle = {Proceedings of the 53rd IEEE/ACM International Symposium on Microarchitecture},
  doi = {10.1109/MICRO50266.2020.00044},
  location = {Virtual Event, Athens, Greece},
  numpages = {15},
  pages = {427--441},
  publisher = {IEEE Press},
  series = {MICRO-53},
  title = {Optimizing the Memory Hierarchy by Compositing Automatic Transformations on Computations and Data},
  url = {https://www.microarch.org/micro53/papers/738300a427.pdf},
  year = {2020},
  bdsk-url-2 = {https://doi.org/10.1109/MICRO50266.2020.00044}
}

2019

TACO 2019

Flextended Tiles: A Flexible Extension of Overlapped Tiles for Polyhedral Compilation

Jie Zhao, and Albert Cohen

ACM Trans. Archit. Code Optim., Dec 2019

Bib PDF

@article{flextile,
  address = {New York, NY, USA},
  articleno = {Article 47},
  author = {Zhao, Jie and Cohen, Albert},
  doi = {10.1145/3369382},
  issn = {1544-3566},
  issue_date = {January 2020},
  journal = {ACM Trans. Archit. Code Optim.},
  keywords = {stencil computations, Polyhedral compilation, automatic parallelization, loop tiling},
  month = dec,
  number = {4},
  numpages = {25},
  publisher = {Association for Computing Machinery},
  title = {Flextended Tiles: A Flexible Extension of Overlapped Tiles for Polyhedral Compilation},
  url = {https://doi.org/10.1145/3369382},
  volume = {16},
  year = {2019},
  bdsk-url-1 = {https://doi.org/10.1145/3369382}
}

ICS 2019

WCCV: Improving the Vectorization of IF-Statements with Warp-Coherent Conditions

Huihui Sun, Florian Fey, Jie Zhao, and Sergei Gorlatch

In Proceedings of the ACM International Conference on Supercomputing, Phoenix, Arizona, Dec 2019

Bib PDF Slides

@inproceedings{wccv,
  author = {Sun, Huihui and Fey, Florian and Zhao, Jie and Gorlatch, Sergei},
  title = {WCCV: Improving the Vectorization of IF-Statements with Warp-Coherent Conditions},
  year = {2019},
  isbn = {9781450360791},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3330345.3331059},
  doi = {10.1145/3330345.3331059},
  booktitle = {Proceedings of the ACM International Conference on Supercomputing},
  pages = {319--329},
  numpages = {11},
  keywords = {IF-statements, compiler optimization, SPMD-on-SIMD, vectorization, warp-coherence},
  location = {Phoenix, Arizona},
  series = {ICS'19}
}

2018

CC 2018

A Polyhedral Compilation Framework for Loops with Dynamic Data-Dependent Bounds

Jie Zhao, Michael Kruse, and Albert Cohen

In Proceedings of the 27th International Conference on Compiler Construction, Vienna, Austria, Dec 2018

Bib PDF Slides

@inproceedings{polydcl,
  author = {Zhao, Jie and Kruse, Michael and Cohen, Albert},
  title = {A Polyhedral Compilation Framework for Loops with Dynamic Data-Dependent Bounds},
  year = {2018},
  isbn = {9781450356442},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3178372.3179509},
  doi = {10.1145/3178372.3179509},
  booktitle = {Proceedings of the 27th International Conference on Compiler Construction},
  pages = {14-24},
  numpages = {11},
  keywords = {parallelizing compiler, loop nest optimization, polyhedral model, dynamic counted loop},
  location = {Vienna, Austria},
  series = {CC 2018}
}