My research interests focus on high performance parallel algorithm design for distributed deep learning, and kernel optimization of neural networks.

(1) September 2007 -- July 2012, Academy of Mathematics and Systems Science, Chinese Academy of Sciences

Ph.D. in Computational Mathematics

(2) September 2003 -- July 2007, Xiangtan University

B.S. in Mathematics

   

(1) September 2018 -- Now, Institute of Computing Technology, Chinese Academy of Sciences

Associate Professor

(2) April 2015 -- September 2018, Institute of Computing Technology, Chinese Academy of Sciences

Assistant Professor

(3) July 2012 -- April 2015, National Astronomical Observatories, Chinese Academy of Sciences

Postdoctor

JOURNAL ARTICLES

(1)  Jiashu Yao, Junmin Xiao, Baokang Xie, Shilong Xu, Xi Chen, Yunfei Pang, Mingyi Li, Hui Ma, Yun Song, and Guangming Tan, "Hiperti: High Performance System for Cross‑Platform Code Generation of Transformer Model Inference Based on MLIR", CCF Transactions on High Performance Computing, Springer, 2025, 7(6). pp. 589--622.

(2) Xun Wang, Ruibao Song, Junmin Xiao, Tong Li, and Xueqi Li, "Accelerating k-Shape Time Series Clustering Algorithm Using GPU", IEEE Transactions on Parallel and Distributed Systems, 2023, 34(10), pp. 2718--2734.

(3) Hang Cao, Liang Yuan, He Zhang, Yunquan Zhang, Baodong Wu, Kun Li, Shigang Li, Minghua Zhang, Pengqi Lu, and Junmin Xiao, "AGCM-3DLF: Accelerating Atmospheric General Circulation Model via 3-D Parallelization and Leap-Format", IEEE Transactions on Parallel and Distributed Systems, 2023, 34(3), pp. 766--780.

(4) Di Cai, Xuehai Hong, Junmin Xiao, and Guangming Tan, "Parallel Optimization for Large-Scale Ocean Data Assimilation", Journal of Computer Research and Development, 2023, 60(5), pp. 1177--1190.

(5) Zhongzhe Hu, Junmin Xiao, Ninghui Sun, and Guangming Tan, "Fast and Accurate Variable Batch Size Convolution Neural Network Training on Large Scale Distributed Systems", Concurrency and Computation: Practice and Experience, Wiley, 2022, 34(21), e7119.

(6) Junmin Xiao, and Jian Peng, "Tradeoffs between Computation, Communication, and Synchronization in Stencil-collective Alternate Update", CCF Transactions on High Performance Computing, Springer, 2019, 1(2), pp. 144--160.

(7) Junmin Xiao, Guizhao Zhang, Yanan Gao, Xuehai Hong, and Guangming Tan, "Fast Data-obtaining Algorithm for Data Assimilation with Large Data Set", International Journal of Parallel Programming, Springer, 2019, 47, pp. 1--21.

(8) Junmin Xiao, Jun Zhang, Ting Li, and Shuhong Yang, "Dark Ribbons Propagating and Sweeping Across Extreme Ultraviolet Structures after Filament Eruptions", The Astrophysical Journal, 2015, 805(1), pp. 25--37.

(9) Huadong Chen, Jun Zhang, Suli Ma, Shuhong Yang, Leping Li, Xin Huang, and Junmin Xiao, "Confined Flares in Solar Active Region 12192 from 2014 October 18 to 29", The Astrophysical Journal Letters, 2015, 808, pp. L24--L31.

(10) Junmin Xiao, and Qiya Hu, "Multilevel Correction for Collocation Solutions of Volterra Integral Equations with Proportional Delays", Advances in Computational Mathematics, 2013, 39(3-4), pp. 611--644.

(11) Xingding Chen, Qiya Hu, and Junmin Xiao, "On the Enhanced Strain Finite Element Method for Incompressible Linear Elasticity", Applied Numerical Mathematics, 2013, 72, pp. 131--142.

CONFERENCE PROCEEDINGS

(1) Yaoyu Wang, Hankun Dai, Zhidong Yang, Junmin Xiao, and Guangming Tan, "Mastering Sparse CUDA Generation through Pretrained Models and Deep Reinforcement Learning", The 14th International Conference on Learning Representations (ICLR 2026), Rio de Janeiro, Brazil, 2026-04-23~2026-04-27.

(2) Zehua Wang, Junmin Xiao, Xiaochuan Deng, Huibing Wang, Hui Ma, Mingyi Li, Yunfei Pang, and Guangming Tan, "T-Control: An Efficient Dynamic Tensor Rematerialization System for DNN Training", The 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2026), Pittsburgh, PA, USA, 2026-03-22~2026-03-26.

(3) Mingyi Li, Junmin Xiao, Siyan Chen, Hui Ma, Xi Chen, Peihua Bao, Liang Yuan, and Guangming Tan, "Stencil-Lifting: Hierarchical Recursive Lifting System for Extracting Summary of Stencil Kernel in Legacy Codes", The ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications (OOPSLA 2025), Singapore, 2025-10-12~2025-10-18. 

(4) Yaoyu Wang, Xiao Guo, Junmin Xiao, De Chen, and Guangming Tan, "GeneralSparse: Bridging the Gap in SpMM for Pruned Large Language Model Inference on GPUs", The USENIX Conference on Usenix Annual Technical Conference (USENIX ATC 2025), Boston, MA, USA, 2025-07-07~2025-07-09.

(5) Mingyi Li, Junmin Xiao, Kewei Zhang, Zhiheng Lin, Chaoyang Shui, Ke Meng, Zehua Wang, Yunfei Pang, and Guangming Tan, "A Coordinated Strategy for GNN Combining Computational Graph and Operator Optimizations", The 38th ACM International Conference on Supercomputing (ICS 2024), Kyoto, Japan, 2024-06-05~2024-06-07.

(6) Zhiheng Lin, Ke Meng, Chaoyang Shui, Kewei Zhang, Junmin Xiao, and Guangming Tan, "Exploiting Fine-Grained Redundancy in Set-Centric Graph Pattern Mining", The 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming (PPoPP 2024), Edinburgh, United Kingdom, 2024-03-02~2024-03-06.

(7) Junmin Xiao, Chaoyang Shui, Di Cai, Kangyu Wang, Yunfei Pang, Mingyi Li, Hui Ma, and Guangming Tan, "Adaptive Workload-Balanced Scheduling Strategy for Global Ocean Data Assimilation on Massive GPUs", The International Conference for High Performance Computing, Networking, Storage and Analysis (SC 2023), Denver, CO, USA, 2023-11-12~2023-11-17.

(8) Kewei Zhang, Junmin Xiao, Zhiheng Lin, Ke Meng, Chaoyang Shui, Mingyi Li, and Guangming Tan, "GraphPar: Efficient Workload-Aware Subgraph Matching System on Multiple GPUs", The IEEE 29th International Conference on Parallel and Distributed Systems (ICPADS 2023), Ocean Flower Island, Hainan, China, 2023-12-17~2023-12-21.

(9) Junmin Xiao, Yunfei Pang, Qing Xue, Chaoyang Shui, Ke Meng, Hui Ma, Mingyi Li, Xiaoyang Zhang, and Guangming Tan, "W-Cycle SVD: A Multilevel Algorithm for Batched SVD on GPUs", The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC 2022), Dallas, Texas, USA, 2022-11-13~2022-11-18.

(10) Zhongzhe Hu, Junmin Xiao, Zheye Deng, Mingyi Li, Kewei Zhang, Xiaoyang Zhang, Ke Meng, Ninghui Sun, and Guangming Tan, "MegTaiChi: Dynamic Tensor-based Memory Management Optimization for DNN Training", The 36th ACM International Conference on Supercomputing (ICS 2022), Virtual Event, USA, 2022-06-28~2022-06-30.

(11) Junmin Xiao, Qing Xue, Hui Ma, Xiaoyang Zhang, and Guangming Tan, "POSTER: A W-cycle Algorithm for Efficient Batched SVD on GPUs", The 27th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP 2022), Seoul, Republic of Korea,  2022-04-02~2022-04-06. 

(12) Xiaoyang Zhang, Junmin Xiao, and Guangming Tan, "I/O Lower Bounds for Auto-tuning of Convolutions in CNNs", The 26th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP 2021), Virtual Event, Republic of Korea, 2021-02-27~2021-03-03. 

(13) Xiaoyang Zhang, Junmin Xiao, and Guangming Tan, "Brief Announcement: Communication Lower Bounds of Convolutions in CNNs", The 32nd ACM Symposium on Parallelism in Algorithms and Architectures (SPAA 2020), Virtual Event, USA, 2020-07-15~2020-07-17. 

(14) Junmin Xiao, Shijie Wang, Weiqiang Wan, Xuehai Hong, and Guangming Tan, "S-EnKF: Co-designing for Scalable Ensemble Kalman Filter", The 24th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP 2019), Washington, D.C., USA, 2019-02-16~2019-02-20. 

(15) Guizhao Zhang, Junmin Xiao, Xuehai Hong, and Guangming Tan, "Fast Data-obtaining Algorithm for Data Assimilation with Large Data Set", The 16th International Conference on Network and Parallel Computing (NPC 2019), Hohhot, Inner Mongolia, China, 2019-08-23~2019-08-24.

(16) Zhongzhe Hu, Junmin Xiao, Zhongbo Tian, Xiaoyang Zhang, Chengji Yao, Ninghui Sun, and Guangming Tan, "A Variable Batch Size Strategy for Large Scale Distributed DNN Training", The 17th IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA 2019), Fujian, Xiamen, China, 2019-12-16~2019-12-19 (BEST PAPER AWARD).

(17) Xiaoyang Zhang, Junmin Xiao, Xiaobin Zhang, Zhongzhe Hu, Hongrui Zhu, Zhongbo Tian, and Guangming Tan, "Tensor Layout Optimization of Convolution for Inference on Digital Signal Processor", The 17th IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA 2019), Fujian, Xiamen, China, 2019-12-16~2019-12-19.

(18) Junmin Xiao, Shigang Li, Baodong Wu, He Zhang, Kun Li, Erlin Yao, Yunquan Zhang, and Guangming Tan, "Communication-avoiding for Dynamical Core of Atmospheric General Circulation Model", The 47th ACM International Conference on Parallel Processing (ICPP 2018), Eugene, Oregon, USA, 2018-08-13~2018-08-16.

(19) Baodong Wu, Shigang Li, Hang Cao, Yunquan Zhang, He Zhang, Junmin Xiao, and Minghua Zhang, "AGCM3D: A Highly Scalable Finite-Difference Dynamical Core of Atmospheric General Circulation Model based on 3D Decomposition", The 24th IEEE International Conference on Parallel and Distributed Systems (ICPADS 2018), Sentosa, Singapore, 2018-12-11~2018-12-13.