个人简介：
李洪亮，博士、副教授、博士生导师，任中国计算机学会分布式计算与系统专业委员会执行委员、容错计算专委会执行委员。于2012年在吉林大学计算机学院获得博士学位、曾于2011年在加拿大多伦多大学做访问学者、于2011年在加拿大Platform Computing Inc.任软件工程师，于2016-2017年在美国天普大学吴杰教授课题组做博士后访问学者。曾获吉林省科技进步一等奖1项，中国商业联合会科技进步一等奖1项，所负责的“云计算技术”系列课程获批国家级一流本科课程虚拟仿真课程和吉林省一流本科课程国家，获ACM中国新星奖和IBM优秀教师奖教金等个人奖项。
主要研究方向为并行/分布式深度学习、云计算和数据中心，研究系统资源调度和各类并行作业容错方法等。主持国家自然基金项目等国家级科研项目2项，省级科研项目1项，国际合作项目1项；参与国家级/省级科研项目5项等，相关成果发表论文40余篇(包括ATC、INFOCOM和CLUSTER等顶级会议和TPDS、TACO、JPDC、JCST和FGCS等顶级期刊)，获得国家发明专利8项。
招生信息：
招收计算机相关专业博士和硕士研究生，欢迎对并行/分布式计算、大规模AI基础设施等研究方向感兴趣、具有较好数学和英语基础的同学联系和报考。可联系我的邮箱(lihongliang@jlu.edu.cn)。
团队新闻：
NEWS！(09/2026) Our paper“Parallel DNN Training with Uniform Local Batch in Heterogeneous Environments”is accepted by IEEE Transactions on Parallel and Distributed Systems. (IEEE TPDS), (CCF-A期刊)!
NEWS！(08/2026) Our paper“Low-latency DNN Model Exploration via Opportunistic Resource Sharing”is accepted by IEEE Transactions on Parallel and Distributed Systems. (IEEE TPDS), (CCF-A期刊)!
NEWS！(06/2026) Our paper“Sift: Channel-Wise Historical Embedding for High Efficiency Distributed Graph Neural Network Training with Accuracy Guarantee”is published by ACM Transactions on Architecture and Code Optimization (ACM TACO), (CCF-A期刊)!
NEWS！(12/2025) Our paper“Rehabilitating over Recomputing: A Novel Failure Recovery Method for Large Model Training”is accepted by IEEE International Conference on Computer Communications (IEEE INFOCOM 2026), (CCF-A会议)!
NEWS！(05/2025) Our paper“FlexPipe: Maximizing the Training Efficiency for Transformer-based models with Variable-Length Inputs”is accepted by USENIX Annual Technical Conference (USENIX ATC2025), (CCF-A会议、系统结构顶级会、吉林大学首篇)!
NEWS！(03/2025) Our paper“Harnessing dynamic graph differential operators for efficient data-driven wind prediction”is accepted byGeoInfomatica, (CCF-B期刊)!
NEWS！(02/2025) Our paper“Alleviating Straggler Impacts for Data Parallel Deep Learning with Hybrid Parameter Update”is accepted by Future Generation Computer Systems (FGCS), (中科院一区)!
NEWS！(12/2024) Our paper“ArrayPipe: Introducing Job-Array Pipeline Parallelism for High Throughput Model Exploration.”is accepted by IEEE International Conference on Computer Communications (IEEE INFOCOM 2025), (CCF-A会议)!
NEWS！(11/2024) Our paper“Convergence-aware Optimal Checkpointing for Exploratory Deep Learning Training Jobs”is accepted by Future Generation Computer Systems (FGCS), (中科院一区)!
NEWS！(11/2024) Our paper“Visage: Visual-Aware Generation of Adversarial Examples in Black-Box for Text Classification”is awarded theBest Paper AwardinNLPCC2024, (CCF-C会议)!
讲授课程：
《云计算技术A》，计算机学院本科生选修课
《云计算技术B》，计算机学院本科生限选课
《云计算技术》，软件学院本科生限选课
《分布式存储》，软件学院本科生必修课
《分布式计算和云计算》，计算机学院研究生选修课
《云计算技术》，软件学院研究生选修课
《云计算技术综合虚拟仿真实验》获国家级一流本科课程
教育和工作经历：
2016至今，吉林大学，计算机科学与技术学院，副教授
2016-2017，美国，天普大学，博士后访问学者
2012-2016，吉林大学，计算机科学与技术学院，讲师
2011-2011，加拿大，Platform Computing Inc.，软件工程师
2011-2011，加拿大，多伦多大学，访问学者
2009-2012，吉林大学，计算机科学与技术学院，博士
2006-2009，吉林大学，计算机科学与技术学院，硕士
2002-2006，吉林大学，计算机科学与技术学院，本科
论文成果：
Selected Publications
Journal Papers
[1]Q. Tian,H. Li*, H. Zhao, Z. Wang, X. Liu, Z. Xu, J. Xiao, G. Tan, D. Tao. Parallel DNN Training with Uniform Local Batch in Heterogeneous Environments, IEEE Transactions on Parallel and Distributed Systems. (TPDS) 2026.(CCF-A期刊)
[2]H. Zhao,H. Li*,Qi Tian, H. Xu, Z. Chen, G. Tan, D. Tao. Low-latency DNN Model Exploration via Opportunistic Resource Sharing, IEEE Transactions on Parallel and Distributed Systems. (TPDS) 2026.(CCF-A期刊)
[3]Z. Xu, J. Han, X. Wei,H. Li*, J. Hao, D. Tian, Z. Li, H. Yue. Sift: Channel-Wise Historical Embedding for High Efficiency Distributed Graph Neural Network Training with Accuracy Guarantee. ACM Transactions on Architecture and Code Optimization (TACO). Vol. 23, Non. 2, Jun. 2026.(CCF-A期刊)DOI:10.1145/3809161
[4]H. Li*, Q. Tian, D. Xu, H. Zhao, Z. Xu. Alleviating Straggler Impacts for Data Parallel Deep Learning with Hybrid Parameter Update. Future Generation Computer Systems (FGCS), Feb. 2025.(中科院一区)DOI:10.1016/j.future.2025.107775
[5]H. Li, Z. Wang, H. Zhao, M. Zhang, X. Li, H. Xu. Convergence-aware Optimal Checkpointing for Exploratory Deep Learning Training Jobs. Future Generation Computer Systems (FGCS), Nov. Mar. 2025.(中科院一区)DOI:10.1016/j.future.2024.107597
[6]Z. Xu, B. Pan, X. Wei,H. Li*, D. Tian, Z. Li, C. Liu. HiHa: Introducing Hierarchical Harmonic Decomposition to Implicit Neural Compression for Atmospheric Data.Computers and Geosciences. 2025. (中科院二区) DOI: 10.1016/j.cageo.2025.106078
[7]H. Li, Z. Xu, Z. Fang, N. Zhang, C. Liu*, R. Zhou, X. Wei. MJOFormer: An AdaptiveLand-Ocean Spatio-TemporalTransformerfor Madden Julian OscillationForecasting.Computers and GeoSciences. 2025.(中科院二区)DOI:10.1016/j.cageo.2025.106097
[8]X. Wei, Z. Xu,H. Li*, J. Hao, H. Yue, C. Liu. Harnessing dynamic graph differential operators for efficient data-driven wind predictionGeoInformatica, Mar. 2025. (CCF-B期刊) DOI:10.1007/s10707-025-00542-2
[9]Z. Xu, B. Pan, X. Wei*,H. Li*, D. Tian, Z. Li. GDOSphere: A Spherical Graph Neural Network Framework with Neural Operators for Weather Forecasting. Physica A, Jun. 2025. DOI: 10.1016/j.physa.2025.130772(中科院二区)
[10]H. Li, H. Zhao, T. Sun, X. Li, H. Xu, K. Li. Interference-aware Opportunistic Job Placement for Shared Distributed Deep Learning Clusters. Journal of Parallel and Distributed Computing (JPDC), Jan. 2024. (CCF-B期刊)DOI:10.1016/j.jpdc.2023.104776
[11]Z. Xu, X Wei, J Hao, J Han,H Li*, C L, Z Li, D Tian, N Zhang. DGFormer: A Physics-Guided Station Level Weather Forecasting Model with Dynamic Spatial-Temporal Graph Neural Network,GeoInformatica, Feb. 2024.(CCF-B期刊)DOI:10.1007/s10707-024-00511-1
[12]Z. Xu, X, Wei, J. Hao, J. Li,H. Li*, Z. Ding, S. Li. HiRM: Hierarchical Resource Management for Earth System Models on Many-core Clusters. CCF Transactions on High Performance Computing (THPC). Jan. 2024.(CCF-C期刊)DOI:10.1007/s42514-023-00176-6
[13]H. Li, J. Wu, Z. Jiang, X. Li, X. Wei. A Task Allocation Method for Stream Processing with Recovery Latency Guarantee. Journal of Computer Science and technology (JCST), vol.33, no.6, pp.1125-1139, 2018.11.(CCF-B期刊)DOI:10.1007/s11390-018-1876-6
[14]H. Li, J. Wu, Z. Jiang, X. Li, X. Wei. Minimum Backups for Stream Processing with Recovery Latency Guarantees.IEEE Transactions on Reliability,vol.66, no.99, pp.1-12. 2017.(中科院二区)DOI:10.1109/TR.2017.2712563
[15]X. Wei, L. Li, X. Li, X. Wang, S. Gao.H. Li. Pec: Proactive Elastic Collaborative Resource Scheduling in Data Stream Processing. IEEE Transactions on Parallel and Distributed Systems (TPDS), vol. 30. No. 7, pp. 1628-1642, July 1 2019.(CCF-A期刊)DOI:10.1109/TPDS.2019.2891587
[16]X. Wei, Z. Xu,H. Li, Z. Ding. Coordinated process scheduling algorithms for coupled earth system models. Concurrency and Computation: Practice and Experience (CCPE), e6346, Oct 25, 2021.(CCF-C期刊)
[17]Y. Zhuang, X. Wei,H. Li, Y. Wang, X. He. An optimal checkpointing model with online OCI adjustment for stream processing applications. Concurrency and Computation: Practice and Experience (CCPE). June 10, 2019.(CCF-C期刊)DOI:10.1002/cpe.5347
[18]X. Wei, Y. Zhuang,H. Li, Z. Liu. Reliable stream data processing for elastic distributed stream processing systems. Cluster Computing. May 21, 2019.(中科院三区)DOI:10.1007/s10586-019-02939-9
[19]W. Wei, X. Wei,H. Li. Topology-aware Task Allocation for Online Distributed Stream Processing Applications with Latency Constraints. Physica A: Statistical Mechanics and its Applications. Vol. 534, Nov. 15, 2019.(中科院二区)DOI:10.1016/j.physa.2019.122024
[20]X. Wei,H. Li, K. Yang, L. Zou. Topology-aware Partial Virtual Cluster Mapping Algorithm on Shared Distributed Infrastructures. IEEE Transactions of Parallel and Distributed Systems (TPDS), vol.25, no.10, pp.2721-2730, October 2014.(CCF-A期刊)DOI:10.1109/TPDS.2013.224
[21]H. Li, X. Wei, Q. Fu, Y. Luo. MapReduce Delay Scheduling with Deadline Constraint. Concurrency and Computation: Practice and Experience (CCPE), vol.26, no.3, pp.766-778, March 10, 2014.(CCF-C期刊)DOI:10.1002/cpe.3050
[22]X. Wei, Y. Jin,H. Li, X. Wang and S. Hu. Virtual Resource Consolidation for Green Computing Based on Virtual Cluster Live Migration. Journal of Communications, vol.11, no.2, pp.192-202, February 2016. DOI:10.12720/jcm.11.2.192-202
[23]X. Wei, W. Li, H. Tian,H. Li, H. Xu, T. Xu. THC-MP: High Performance Numerical Simulation of Reactive Transport and Multiphase Flow in Porous Media. Computers & Geosciences, vol. 80, pp.26-37, 2015.
[24]X. Wei, X. Bai, S. Bai andH. Li. On-demand Tile Preload for Large-scale Seismic Data 3D-visualization. Journal of Computational Information Systems, vol.11, no.4, pp.1513-1520, February 2015. DOI:10.12733/jcis13585
[25]X. Wei, S. Hu,H. Li, F. Yang, Y. Jin. A survey on virtual network embedding in cloud computing centers. Open Automation and Control Systems Journal, vol.6, no.1, pp.414-425, 2014.
[26]X. Wei,H. Li, L. Hu, Q. Guo, N. Jiang. LimeVI: A Platform for Virtual Cluster Live Migration over WAN. International Journal of Computer Systems Science and Engineering (CSSE), vol.26, No.5, pp.353-364, September 2011.
Conference Papers
[1]Z. Wang, H. Li*, J. Wu, Z. Xu, H. Zhao, Q. Tian, H. Xu. Rehabilitating over Recomputing: A Novel Failure Recovery Method for Large Model Training. IEEE International Conference on Computer Communications (INFOCOM2026), May 2026.(CCF-A会议)
[2]H. Zhao, Q. Tian,H. Li*, Z. Chen. FlexPipe: Maximizing the Training Efficiency for Transformer-based models with Variable-Length Inputs. USENIX Annual Technical Conference (USENIX ATC2025), Jul. 07-09, 2025. (CCF-A会议,吉林省/吉林大学首篇)
[3]H. Zhao,H. Li*, Q. Tian, J. Wu, M. Zhang, Z. Xu, X. Li, H. Xu. ArrayPipe: Introducing Job-Array Pipeline Parallelism for High Throughput Model Exploration. IEEE International Conference on Computer Communications (INFOCOM2025), May, 2025.(CCF-A会议)
[4]H. Li, H. Zhao, Z. Xu, X. Li, and H. Xu. ExplSched: Maximizing Deep Learning Cluster Efficiency for Exploratory Jobs. IEEE International Conference on Cluster Computing (CLUSTER2023), Oct. 31, 2023, Santa Fe, New Mexico, USA.(CCF-B会议)
[5]H. Zhao, X. Li,H. Li*. Visage: Visual-Aware Generation of Adversarial Examples in Black-Box for Text Classification. The 13th CCF International Conference on Natural Language Processing and Chinese Computing (NLPCC 2024), Nov. 1, 2024, Hangzhou, China.(the Best Paper Award) (CCF-C会议)
[6]H. Li, D. Xu, Z. Xu, X. Li. Hybrid Parameter Update: Alleviating Imbalance Impacts for Distributed Deep Learning. 24th IEEE International Conference on High Performance Computing and Communications (HPCC2022), Dec. 2022.(CCF-C会议)
[7]H. Li, T. Sun, X. Li, H. Xu. Job Placement Strategy with Opportunistic Resource Sharing for Distributed Deep Learning Clusters. 2020 IEEE 22nd International Conference on High Performance Computing and Communications (HPCC2020), Dec. 2020.(CCF-C会议)
[8]H. Li, Z Xu, F. Tang, X. Wei, Z. Ding. CPSA: A Coordinated Process Scheduling Algorithm for Coupled Earth System Model. 2020 29th International Conference on Computer Communication and Networks (ICCCN), August 2020.(CCF-C会议)
[9]Y. Zhuang, X. Wei,H. Li*, M. Hou, Y. Wang. Reducing Fault-tolerant Overhead for Distributed Stream Processing with Approximate Backup. 2020 29th International Conference on Computer Communication and Networks (ICCCN), August 2020.(CCF-C会议)
[10]Y. Zhuang, X. Wei,H. Li*, Y. Wang and X. He. An Optimal Checkpointing Model with Online OCI Adjustment for Stream Processing Applications. 2018 27th International Conference on Computer Communication and Networks (ICCCN), pp. 1-9, July 30 2018, Hangzhou, China.(CCF-C会议)DOI:10.1109/ICCCN.2018.8487327
[11]H. Li, J. Wu, Z. Jiang, X. Li, X. Wei. Task Allocation for Stream Processing with Recovery Latency Guarantee. in Cluster Computing (CLUSTER), 2017 IEEE International Conference on. IEEE, 2017, pp. 379–383.(CCF-B会议)DOI:10.1109/CLUSTER.2017.10
[12]H. Li, J. Wu, Z. Jiang, X. Li, X. Wei, Y. Zhuang. Integrated Recovery and Task Allocation for Stream Processing. 2017 IEEE 36th International Performance Computing and Communications Conference (IPCCC), Dec. 10, 2017, San Diego, CA, USA.(CCF-C会议)DOI:10.1109/PCCC.2017.8280443
获奖情况：
国家级一流本科课程虚拟仿真课程2023
吉林省科技进步一等奖2018
中国商业联合会科学技术奖一等奖2014
CSC-IBM中国优秀教师奖教金2014
ACM中国新星奖2015
HPC China 2015会议优秀论文奖2015
联系方式：
lihongliang@jlu.edu.cn
