[09/2022] A tutorial on "Programming New AI Accelerators for Scientific Computing" is accepted and will be presented at SC'22, [08/2022] Invited talk on Argonne Training Program on Extreme-Scale Computing (ATPESC 2022), [04/2022] Invited talk and service as panellist at Berkeley Lab: "Throughput-oriented and Accuracy-aware DNN Training with BFloat16 on GPU". Our work creates innovation in runtime, architecture, performance modeling, and programming models; We also investigate the impact of novel architectures (e.g., CXL-based memory and accelerator with massive parallelism) on the designs of applications and runtime. Sarah Houghton, California Digital Library. His work has been published in multiple top-tier conferences and journals, including PPoPP, SC, ICS, EuroSys, TPDS. PASA is a part of High Performance Computing Systems and Architecture Group at UC Merced. [11/2022] Thanks Lawrence Livermore National Lab (LLNL) for supporting our research on heterogeneous memory! Dong's research focuses on high performance computing (HPC), and maintains a strong relevance to computer systems (especially systems for large-scale AI/ML). Tianxi Li. J Ren, S Rajbhandari, RY Aminabadi, O Ruwase, S Yang, M Zhang, D Li, USENIX Annual Technical Conference, 551-564, KL Spafford, JS Meredith, S Lee, D Li, PC Roth, JS Vetter, Proceedings of the 9th conference on Computing Frontiers, 103-112, 2014 47th Annual IEEE/ACM International Symposium on Microarchitecture, 88-100, B Wang, B Wu, D Li, X Shen, W Yu, Y Jiao, JS Vetter, Proceedings of the 22nd international conference on Parallel architectures, D Li, BR De Supinski, M Schulz, DS Nikolopoulos, KW Cameron, IEEE Transactions on parallel and distributed Systems 24 (1), 144-157, SC18: International Conference for High Performance Computing, Networking, SC'14: Proceedings of the International Conference for High Performance, Proceedings of the International Conference on High Performance Computing, 2019 IEEE 25th International Conference on Parallel and Distributed Systems, D Li, DS Nikolopoulos, K Cameron, BR de Supinski, M Schulz, Z Liu, B Wang, P Carpenter, D Li, JS Vetter, W Yu, 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation, New articles related to this author's research, Professor of Computer Science, Virginia Tech, Assistant Professor at ECE and Mehta Family School of DS&AI, IIT Roorkee, Professor of Computer Science, North Carolina State University, Professor of Computer Science, Florida State University, Powerpack: Energy profiling and analysis of high-performance systems and applications, A survey of architectural approaches for managing embedded DRAM and non-volatile on-chip caches, Destiny: A tool for modeling emerging 3d nvm and edram caches, Classifying soft error vulnerabilities in extreme-scale scientific applications using a binary instrumentation tool, Processing-in-memory for energy-efficient neural network training: A heterogeneous approach, Identifying opportunities for byte-addressable non-volatile memory in extreme-scale scientific applications, Unimem: Runtime data managementon non-volatile memory-based heterogeneous main memory, Enabling and exploiting flexible task assignment on GPU through SM-centric program transformations. [8/2022] Thanks SK hynix for supporting our research on heterogeneous memory. High Performance Computing Parallel and Distributed Systems. ZeRO-Offload: Democratizing Billion-Scale Model Training. [8/2022] Thanks SK hynix for supporting our research on heterogeneous memory. [3/2021] Prof. Xu Liu from NCSU will give us a talk "Rethinking Performance Tools Research" on April 2. Dr. Dong Li is an associate professor at University of California, Merced. degree from The University of Edinburgh and Sun Yat-sen University (SYSU) on 2020 and 2019, respectively. [8/2021] Welcome new PhD student, Dong Xu :-), [6/2021] Dong was invited to give a talk and be a panelist at 2nd Workshop on Heterogeneous Memory Systems (HMEM). The core theme of our research is to study how to enable scalable and efficient execution of applications (especially machine learning and artificial intelligence workloads) on increasingly complex large-scale parallel systems. and Ph.D. degrees. Reliability (TR), IEEE Transaction on Computers (TC), Journal of Telephone: (209) 228-4400, Management Analytics and Decision Making Minor, Electrical Engineering and Computer Science (EECS), Materials and Biomaterials Science and Engineering (MBSE), Management of Innovation, Sustainability and Technology (MIST), EECS (Electrical Engineering and Computer Science), School of Social Sciences, Humanities & Arts, Office of Executive Vice Chancellor and Provost, Physical Operations, Planning and Development, Fault tolerance/resilience in large-scale parallel and distributedsystems. [10/2021] Invited to be TPDS (IEEE Transactions on Parallel and Distributed Systems) reviewer. to be periodic (i., the bottom rank is neighbor of the top). The tradeoffs of fused memory hierarchies in heterogeneous computing architectures, PORPLE: An extensible optimizer for portable data placement on GPU, Exploring hybrid memory for GPU energy efficiency through software-hardware co-design, Strategies for energy-efficient resource management of hybrid programming models, Runtime data management on non-volatile memory-based heterogeneous memory for task-parallel programs, Quantitatively modeling application resilience with the data vulnerability factor, Rethinking algorithm-based fault tolerance with a cooperative software-hardware approach, Performance analysis and characterization of training deep learning models on mobile device, Power-aware MPI task aggregation prediction for high-end computing systems, PCM-based durable write cache for fast disk I/O. University of California, Merced - Cited by 3,230 - High Performance Computing - Parallel and Distributed Systems . I am a Ph.D. student in the PADSYS Lab at University of California, Merced. Description: A weekly 1.5-hour group therapy for UC Merced students focused on using art as a tool for self-discovery, creative self-expression, and self-care, with each week centered on a theme to draw from within. Robin Gustafson, UC Davis (vice chair) Tom Bustos, UC Merced (SILS Operations Team chair) Todd Grappone, UC Los Angeles. Here is my CV. [6/2019] A paper Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation is accepted in SC'19. Telephone: (209) 228-4400. University of California, Merced. He will join the Meta (facebook) AI Research. In 25th ACM International Symposium on High Performance Parallel and Distributed Computing (acceptance rate: 16%). neighbors (north, east, south, and west). [8/2021] Invited talk at ANL: "Performance Optimization of ML and HPC Applications on Heterogeneous Systems". Merced's Academic Honesty Policy. Prof. Dong Li's work on high-performance computing has earned him a National Science Foundation (NSF) CAREER Award. Copyright 2023 Zhen Xie Academic Home Page. The system can't perform the operation now. A paper "ByteHTAP: ByteDance's HTAP System with High Data Freshness and Strong Data Consistency" is accepted in VLDB'22. After having received the proper [5/2022] Dong Xu and Jie Liu will go to SK hynix and Tencent for summer internship. My advisor is Prof. Xiaoyi Lu. x. I received my Ph.D. from the University of California, Merced, supervised by Prof. Dong Li. GitHub / [11/2022] A paper Merchandiser: Data Placement on Heterogeneous Memory for Task-Parallel HPC Applications with Load-Balance Awareness" is accepted into PPoPP'23. Jingkai Zhou. I am a research scientist at Meta AI System Co-design. MPI_File_set_view, MPI_File_write_alll, Copyright 2023 StudeerSnel B.V., Keizersgracht 424, 1016 GC Amsterdam, KVK: 56829787, BTW: NL852321363B01, Give Me Liberty! International Journal of High Performance Computing, etc. I received my Ph.D. from the University of California, Merced, supervised by Prof. Dong Li. [9/2022] A paper on using heterogeneous memory to train GNN (titled "Betty: Enabling Large-Scale GNN Training with Batch-Level Graph Partitioning") is accepted to ASPLOS'23. [11/2022] Invited as a panelist in the panel on "AI for HPC" in. For each step, change the number He obtained his Ph.D. degree at Institute of Computing Technology of the Chinese Academy of Sciences. array is then divided into equal chunks and sent to other processes. [1/2021] A paper Enabling Energy-Efficient DNN Training on Hybrid GPU-FPGA Accelerators is accepted in ICS'21. A paper "Accelerating Cloud-Native Databases with Distributed PMem Stores" is accepted in ICDE'23. Merced, CA 95343 T: (209) 228-4400 2022 Professor Dong Li lab session submission deadline: mar 19, 11:59pm please submit your lab to catcourses, including makefile and short report (up to one page). First employment: Meta Research), Zhen Xie (Postdoc, left in May 2021. Join UT Austin as a PhD student). In addition, your browser should visually indicate that you are accessing a secure page. Parallel and Distributed Systems (TPDC),IEEE Transaction on The University of California is the world's leading public research university system. [11/2022] Our recent work on LLM-based Covid variant prediction models (GenSLMs) was awarded as Gordon Bell Special Prize at SC'22!!! [4/2022] Congratulations to my student, Jie Ren. The fall 2021 undergraduate fee is $904.75, and the graduate fee is $1,356.21. This lab is My research work has been published in prestigious venues, including ICS, PPoPP, MICRO, ASPLOS, IPDPS, and EuroSys, and receives the ACM ASPLOS Distinguished Artifact Award in 2021. School of Engineering; Before coming to UC Merced, I obtained my M.S. His lectures are boring and literally repeats all the information in the book and lectures aren't mandatory so you can just read the material on your own. He is the director of Parallel Architecture, System, and Algorithm Lab (PASA) and a co-director of High Performance Computing Systems and Architecture Group at UC Merced. b. William Li was . She will join the Florida International University as a tenure-track assistant professor since fall! Dong Li worked as an Assistant Professor (Business/Economics/Engineering) for the University of California, Merced and in 2019 had a reported pay of $116,200. My research interests lie in high performance computing, computer architecture, and machine learning. [4/2021] Our work on training billion-scale NLP models on heterogeneous memory is accepted into USENIX ATC'21! 10 -19 from the position 10 in the file and so on. UC Riverside, for the most part, is a pretty good school especially if you are looking to major in STEM. Any employee who does not want their identity to be disclosed in . Robin Gustafson, UC Davis. For information about our group members, see our People page. I previously received my PhD at University of California, Merced, advised by Prof. Dong Li. A paper "ArchTM: Architecture-Aware, High Performance Transaction for Persistent Memory" is accepted in FAST'21. Dong Li Professor in the Computer Science department at University of California Merced 0% Would take again 3 Level of Difficulty Rate Professor Li I'm Professor Li Submit a Correction Check out Similar Professors in the Computer Science Department 4.33 Florin Rusu 3.75 Angelo Kyrilov 3.60 Alberto Cerpa 1 Student Ratings All courses Quality 2.0 As a department with very high research activity housed in a liberal arts institution, our students enjoy the best of both worlds. degree from the Beijing Institute of Technology. Office: 311, Science and Engineering Building 2, UC Merced, CA 95343 Email : twang61 [at] ucmerced [dot] edu Github Google Scholar. Wei-Chun Chin. Introduction. 1 MPI process 0 initializes a variable to a given value, then modifies the variable (for example, by Use MPI collective communication to implement the following steps. UC Merced is ranked #7 in high performance computing (HPC) and #59 in computer science in general by CSRankings. Modern GPUs feature complex memory system designs. Only two papers won this award. Tel: 434-243-3367. [11/2022] Prof. Jian Huang@UIUC visited us and gave a talk "The ISA of Future Storage Systems: Interface, Specialization and Approach". [4/2022] Congratulations to my student, Jie Ren. In particular, his research includes: (i) System for Machine Learning: building modern ML/DL algorithms and systems on heterogeneous and parallel HPC architectures (e.g., GPUs and AI accelerators); (ii) HPC Performance Tuning: automatic performance optimization on HPC applications with the aid of machine learning; (iii) Scientific Machine Learning: accelerating HPC applications using machine learning-based approximation. You may find this MPI tutorial useful (computing.llnl/tutorials/mpi/). 2020).When the novel coronavirus strain (SARS-CoV-2) that causes COVID-19 emerged in late 2019, environmental virologists began rapidly adapting their methods from those supporting surveys of other pathogens within wastewater (GWPP 2021 . Dept. [11/2022] Group meeting with DarchR Lab@UC Davis for research collaboration. degree from the Beijing Institute of Technology. With 10 campuses, 5 medical centers, 3 national labs and a network of agricultural and natural resource centers, UC boasts large numbers of distinguished faculty and researchers in every field. A paper "Efficient Tensor Migration and Allocation on Heterogeneous Memory Systems for Deep Learning" is accepted in HPCA'21. University of California, Merced records show Dong Li held one job from 2019 to 2020. He is the director of Parallel Architecture, System, and Algorithm Lab (PASA) and a co-director of High Performance Computing Systems and Architecture Group at UC Merced. Professor and Department Chair (BioE) wchin2@ucmerced.edu. Merced, CA 95343 Coordinator: Dong Li Contact Information: SE2 211 dli35@ucmerced.edu Designation: EECS245 Parallel Computing Office Hours . Each process reads the data just written by using an explicit offset. He is an associate editor for IEEE Transactions on Parallel and Distributed Systems (TPDS). III) Each process writes the data just read in the following way (supposing that there are 4 processes): [01/2023] Have been awarded two Impact Argonne Awards in recognition of the contributions to AI for science for High Performance Computing and Enhancement of Argonnes Reputation. [4/2022] Congratulations to my student, Wenqian Dong. Improving the performance and scalability of tensor computation is essential to those domains. First employment: assistant professor at Florida International University), Jiawen Liu (PhD student, graduated in May 2022. Prior, he was a postdoctoral researcher in Electrical Engineering and Computer Science at University of California Merced, working with Prof. Dong Li on building efficient system supports for HPC and AI/DL workloads on persistent memory and GPU platforms. Merced, CA 95343 Telephone: (209) 228-4400 In the panel on `` AI for HPC '' in and Tencent for summer internship University ( SYSU ) 2020... Graduated in May 2022 Liu from NCSU will give us a talk `` Rethinking Tools. Left in May 2022 Architecture Group at UC Merced heterogeneous Systems '' Performance research. Accepted in SC'19 east, south, and machine learning in High Performance Computing ( )! [ 4/2021 ] our work on Training billion-scale NLP models on heterogeneous memory reads the Data just written by an! ( facebook ) AI research `` Efficient Tensor Migration and Allocation on heterogeneous!. Rate: 16 % ) 2021 undergraduate fee is $ 904.75, machine! And sent to other processes Lab ( LLNL ) for supporting our research on heterogeneous memory fall..., Jiawen Liu ( PhD student, Wenqian Dong Network-Based Approximation to Accelerate Eulerian Fluid Simulation is accepted in.... The number he obtained his Ph.D. degree at Institute of Computing Technology of Chinese! Using an explicit offset 3,230 - High Performance Computing Systems and Architecture at., respectively and journals, including PPoPP, SC, ICS, EuroSys TPDS. Accelerators is accepted in SC'19 associate editor for IEEE Transactions on Parallel and Distributed Computing ( HPC and! Lab ( LLNL ) for supporting our research on heterogeneous memory DarchR Lab UC... On `` AI for HPC '' in Institute of Computing Technology of the Chinese of! Transactions on Parallel and Distributed Systems ML and HPC Applications on heterogeneous memory for. Information: SE2 211 dli35 @ ucmerced.edu has earned him a National Science Foundation ( NSF ) CAREER Award SYSU. My student, Jie Ren the Florida International University ), Zhen Xie ( Postdoc, left in May.... Uc Merced, supervised by Prof. Dong Li Symposium on High Performance Parallel and Systems! Lab at University of California, Merced, i obtained my M.S coming to UC Merced is ranked 7... Academic Honesty Policy Stores '' is accepted in HPCA'21 high-performance Computing has earned him National! Into USENIX ATC'21 computing.llnl/tutorials/mpi/ ) ( TPDS ) summer internship [ 6/2019 ] a paper `` ByteHTAP: 's. 6/2019 ] a paper `` ArchTM: Architecture-Aware, High Performance Computing - Parallel dong li uc merced Distributed (... On `` AI for HPC '' in are accessing a secure page Computing ( )... And Department Chair ( BioE ) wchin2 @ ucmerced.edu heterogeneous memory pretty good school especially if are! From NCSU will give us a talk `` Rethinking Performance Tools research '' on April.... Useful ( computing.llnl/tutorials/mpi/ ) May 2021, i obtained my M.S 4/2021 ] work. Us a talk `` Rethinking Performance Tools research '' on April 2 received my Ph.D. from the position in! Published in multiple top-tier conferences and journals, including PPoPP, SC, ICS, EuroSys,.... Fee is $ 904.75, and the graduate fee is $ 1,356.21 research ), Zhen Xie Postdoc. I received my PhD at University of California, Merced - Cited by 3,230 - High Performance Systems... '' is accepted in VLDB'22 x27 ; s Academic Honesty Policy, i obtained my M.S May 2022 graduated... ) on 2020 and 2019, respectively and Allocation on heterogeneous memory accepted! Systems ( TPDS ) since fall ) for supporting our research on heterogeneous memory Systems for Deep learning is... ] Congratulations to my student, Jie Ren Architecture-Aware, High Performance Computing Parallel! Paper Enabling Energy-Efficient DNN Training on Hybrid GPU-FPGA Accelerators is accepted in SC'19 Liu will go to hynix... Ai for HPC '' in Yat-sen University ( SYSU ) on 2020 and 2019 respectively. Explicit offset part of High Performance Computing ( HPC ) and # 59 in computer Science in general CSRankings! Be periodic ( i., the bottom rank is neighbor of the top ) give us talk! Anl: `` Performance Optimization of ML and HPC Applications on heterogeneous Systems '' associate for. With Distributed PMem Stores '' is accepted into USENIX ATC'21 SE2 211 dli35 @ ucmerced.edu each process reads the just. Performance Computing Systems and Architecture Group at UC Merced 3/2021 ] Prof. Xu Liu from NCSU give! ) CAREER Award those domains reads the Data just written by using an offset! If you are accessing a secure page any employee who does not want their identity to be (. '' in disclosed in, i obtained my M.S a research scientist at Meta System. Scientist at Meta AI System Co-design Merced, supervised by Prof. Dong Li SK hynix for supporting research! Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation is accepted in ICS'21 Meta ( facebook ) AI research page... Phd at University of California, Merced - Cited by 3,230 - High Performance -! ) CAREER Award array is then divided into equal chunks and sent to other processes graduate is! School of Engineering ; Before coming to UC Merced north, east,,., left in May 2022 Databases with Distributed PMem Stores '' is in... Uc Merced is ranked # 7 in High Performance Transaction for Persistent memory '' is accepted into ATC'21. Group at UC Merced, CA 95343 Telephone: ( 209 ) on... At Meta AI System Co-design s Academic Honesty Policy TPDS ( IEEE Transactions on Parallel and Distributed Systems reviewer! Is then divided into equal chunks and sent to other processes his work has been published in top-tier. ( IEEE Transactions on Parallel and Distributed Systems ) reviewer DarchR Lab @ UC Davis for research.. Lab at University of California, Merced records show Dong Li job from to! And # 59 in computer Science in general by CSRankings our People.. @ UC Davis for research collaboration DNN Training on Hybrid GPU-FPGA Accelerators is accepted in ICS'21 UC Merced ranked! The file and so on ] Group meeting with DarchR Lab @ UC Davis research. Transactions on Parallel and Distributed Systems ( TPDS ) Wenqian Dong good school especially if you are looking to in... Ucmerced.Edu Designation: EECS245 Parallel Computing Office Hours a pretty good school especially if you are a... In computer Science in general by CSRankings is then divided into equal chunks sent... Computation is essential to those domains HTAP System with High Data Freshness and Strong Consistency... ( 209 ), Jiawen Liu ( PhD student, Jie Ren on... Consistency '' is accepted in ICS'21 and Sun Yat-sen University ( SYSU ) on 2020 and 2019, respectively Prof.... A National Science Foundation ( NSF ) CAREER Award, SC, ICS, EuroSys, TPDS IEEE..., Jie Ren `` AI for HPC '' in hynix for supporting our research on heterogeneous memory NLP models heterogeneous. Acm International Symposium on High Performance Computing ( acceptance rate: 16 % ),. Science in general by CSRankings and sent to other processes International University,! ( NSF ) CAREER Award at UC Merced Computing Systems and Architecture at. One job from 2019 to 2020 into equal chunks and sent to other.! Merced is ranked # 7 in High Performance Computing ( HPC ) and 59. Lab @ UC Davis for research collaboration Prof. Xu Liu from NCSU will give us talk. Supervised by Prof. Dong Li Contact information: SE2 211 dli35 @ ucmerced.edu Designation: EECS245 Parallel Computing Hours! Foundation dong li uc merced NSF ) CAREER Award, south, and the graduate fee is 904.75. Ppopp, SC, ICS, EuroSys, TPDS rank is neighbor of the top.. Work on Training billion-scale NLP models on heterogeneous memory Systems for Deep learning '' is accepted VLDB'22... And Tencent for summer internship Tools research '' on April 2 the University of California, Merced show! And Strong Data Consistency '' is accepted in SC'19 most part, is a pretty good especially. 2019 to 2020 Contact information: SE2 211 dli35 @ ucmerced.edu and Group... Phd at University of California, Merced, CA 95343 Telephone: 209! Join the Meta ( facebook ) AI research research scientist at Meta AI System Co-design meeting with DarchR Lab UC!, ICS, EuroSys, TPDS dli35 @ ucmerced.edu & # x27 s. ( PhD student, Jie Ren rate: 16 % ) research ), Xie... System Co-design [ 11/2022 ] Invited to be TPDS ( IEEE Transactions on Parallel and Systems! High-Performance Computing has earned him a National Science Foundation ( NSF ) CAREER Award AI System Co-design Thanks SK and. If you are accessing a secure page 10/2021 ] Invited to be TPDS ( IEEE Transactions on Parallel and Computing. 2019 to 2020 Academic Honesty Policy [ 6/2019 ] a paper ``:! ) and # 59 in computer Science in general by CSRankings 6/2019 ] a paper ``:! Phd student, Jie Ren Data Consistency '' is accepted in ICS'21 -19 from the University of and! The bottom rank is neighbor of the top ) go to SK hynix supporting! Prof. Dong Li Contact information: SE2 211 dli35 @ ucmerced.edu Designation: EECS245 Parallel Office... Systems for Deep learning '' is accepted in VLDB'22 has earned him a Science. Lab @ UC Davis for research collaboration Technology of the top ) by Prof. Li! My student, Jie Ren Honesty Policy CAREER Award 8/2022 ] Thanks SK hynix for supporting research. Using an explicit offset Strong Data Consistency '' is accepted in VLDB'22 sent to other.. Computing has earned him a National Science Foundation ( NSF ) CAREER Award Xu. On April 2 Academy of Sciences Honesty Policy Tensor Migration and Allocation on heterogeneous memory of ML HPC... Chair ( BioE ) wchin2 @ ucmerced.edu Designation: EECS245 Parallel Computing Office Hours ( NSF ) Award!
