The CCF Distributed Computing Conference (CCF Computility 2024) and the National Conference on Open Distributed and Parallel Computing (DPCS 2024) will be held in Changchun, Jilin Province, China from July 26 to 28, 2024, with an estimated conference size of more than 1,000 people. The theme of this conference is "Computing Network: Distributed System in the Context of New Quality Productivity", which aims to provide the most professional platform for academic discussions, technical exchanges and achievement display for practitioners related to distributed systems and computing power networks. CCF Computility 2024 has prepared 9 keynote reports by academicians and other top experts, and 15 technical forums (80 invited reports), which are definitely not to be missed!
CCF Computility 2024 | 学术盛宴，大咖云集，不容错过！
Keynote speaker of the conference
Background of the Forum
Since the invention of computers, the United States has led the global information technology for more than 70 years, and the hashrate network may become an original invention and project for China's overtaking. Distributed system software is the soul of the computing network, which is responsible for shielding the heterogeneous, heterogeneous, and remote attributes of computing resources, as well as managing diversified computing network applications. This forum will invite scholars and technical experts from all walks of life in the field of computing power network to share and discuss the architecture and abstraction of the system software of the computing power network, the core technology and application requirements of large-scale training and inference for large models, in order to provide basic ideas and guiding ideas for the system software research and development and application ecology construction of the computing power network.
| Arrangement of the report of the software technology forum of the computing network | ||
| Chair of the Forum | Researcher Zhiwei Xu and Associate Researcher Xiaohui Peng | |
| Invited Presentation1 | Chen Haibo is the deputy director of the CCF System Software Committee and a professor at Shanghai Jiao Tong University | Inductive Computation of Low-latency Servers Based on New Operating System Primitives: Characteristics, Optimization, and Reflection |
| Invited Presentation 2 | Liu Qianzhe is a professor at Peking University | System Software in the Era of Large Models: Thinking and Doing |
| Invited Presentation 3 | Hu Chunming is a director of CCF, deputy director of CCF System Software Committee, and professor of Beihang University | undetermined |
| Invited Presentation4 | Zhai Ennan is a member of the Standing Committee of the CCF Internet Committee and the head of Alibaba Cloud Network Research | Heterogeneous computing power abstraction system for computing power network |
| Invited Presentation 5 | Guo Wenchuan, Chief Scientist of China Electronics Co., Ltd., President of CLP Jinxin Research Institute | Practical exploration of computing network technology in the financial industry |
| Invited Presentation6 | Chen Quan is a professor at Shanghai Jiao Tong University | Scheduling and parallel co-design for large-scale model training |
| Time: July 27, 2024 afternoon Location: Chunhua Qiushi Hall, Anhua Holiday Banquet Center, Changchun City, Jilin Province |
Forum Chair & Introduction
Xu Zhiwei
Institute of Computing Technology, Chinese Academy of Sciences
Zhiwei Xu, CCF Fellow, Research Fellow, Institute of Computing Technology, Chinese Academy of Sciences, Professor, University of Chinese Academy of Sciences. He has been supported by the National Fund for Distinguished Young Scholars. He has successively served as Deputy Director, Chief Engineer, Director of Academic Degree Committee, Director of Academic Committee, and Chief Engineer of Shuguang Information Industry Co., Ltd. His main research areas are high-performance computer architecture and distributed system software. He has served as the leader of many important scientific and technological projects such as the pilot project of the Chinese Academy of Sciences, the key research and development program of the Ministry of Science and Technology, and the key project of the National Foundation of China. He has put forward academic ideas such as inclusive computing and human-machine-object ternary computing, and contributed a number of open source software in the fields of big data and cloud computing to the international community, and the research results have been widely used in domestic and foreign enterprises, and won the National Science and Technology Progress Award and the Wang Xuan Award of the China Computer Federation. He has served as a member of the Expert Advisory Committee of the Information Science Department of the National Natural Science Foundation of China, an editorial board member of many international journals such as IEEE TC AND IEEE TSC, and a TPC member of many international conferences such as IEEE HPDC, IEEE HPCA, IEEE SCC, etc., and is currently the editor-in-chief of JCST and Computer Research and Development.
Peng Xiaohui
Institute of Computing Technology, Chinese Academy of Sciences
Xiaohui Peng is an associate researcher at the Institute of Computing Technology, Chinese Academy of Sciences. He is mainly engaged in the research of distributed systems and edge computing, and in recent years, he has focused on the research of information high-speed rail computing power network end-edge-cloud collaborative computing. He has presided over more than 10 longitudinal scientific research projects such as key projects and general projects of the National Science Foundation of China. He has published more than 30 papers in related journals and conferences including PIEEE, IEEE IoT-J, Chinese Journal of Computers, Computer Research and Development, IEEE/ACM SEC, IEEE TKDE, etc. He is also the Associate Editor of IEEE TMC and IEEE IC Journals, ACM/IEEE SEC TPC (2021-2023), a standing member of the Distributed Computing and Systems Committee of the China Computer Federation, and a member of the first Computing Network Committee of the China Institute of Communications.
Presentation of the guests and the presentation
Chen Haibo
Shanghai Jiao Tong University
Haibo Chen, Deputy Director of CCF System Software Committee, Distinguished Professor of Shanghai Jiao Tong University, Chairman of the Technical Steering Committee of OpenHarmony Project Group, Winner of the National Fund for Distinguished Young Scholars, ACM Fellow, IEEE Fellow. His main research areas are operating systems, distributed systems and system security, and the research results have been applied to billions of devices through the in-depth integration of industry, academia and research, and have had a wide range of academic and industrial impacts. He has won the Special Prize of Science and Technology Progress Award of China Computer Federation, Chen Kah Kee Youth Science Award (Information Technology Science Award), China Youth Science and Technology Award, First Prize of Technological Invention Award of the Ministry of Education, and National Excellent Doctoral Dissertation Award. He is currently the first non-North American scholar chair of ACM SIGOPS, the first Chinese scholar editorial board member and field co-chair of ACM's flagship journal, Communications of the ACM, and the co-chair of the ACM EuroSys 2025 program committee. His research work has also won the Huawei Outstanding Contribution Individual Award, the Best Paper Award for SOSP, ASPLOS, EuroSys, and VEE, as well as the DSN "Time Test Award" and the SIGMOD Research Highlight Award. According to csrankings.org statistics, the number of papers published in SOSP/OSDI, the top conference in the field of operating systems in the past 5 years (2019~2023), ranks first in the world. "Modern Operating Systems: Principles and Implementations" won the 2020 "Most Popular IT Book Award" and the 2022 Shanghai Jiao Tong University Best Undergraduate Textbook Award (the new edition was renamed "Operating Systems: Principles and Implementation"), as well as the "Top Ten Best Books in the Computer Field in 2023", which are widely used by universities, scientific research institutions and enterprises.
Title: Low Latency Server Senseless Computation Based on New Operating System Primitives: Features, Optimization, and Reflection
Report Summary: Server Senseless Computing Promises Efficient Software Development and Deployment. To do this, server-imperceptible computing platforms need to address two challenges: strong isolation between function instances and extremely low startup latency. In this report, I will begin with a characterization of the industry's leading server-imperceptible computing platform and derive a set of key metrics that together form a systems approach and a set of benchmarks called Severlessbench (v1 and v2). Then, I'll show how to use novel operating system primitives to make server-agnostic platforms more efficient in normal and confidential server-imperceptible computing on CPU and CPU-XPU platforms. Finally, I will reflect on the gap between server imperceptible computing research and industrial systems, and look forward to the future of server imperceptible computing.
Liu Xunzhe
Peking university
Liu Qianzhe is a Boya Distinguished Professor of Peking University, a recipient of the National Science Foundation for Distinguished Young Scholars, and an ACM Distinguished Scientist. His main research interests are system software, cloud computing, service computing, etc. He has published more than 80 papers in conferences and journals such as SOSP, OSDI, NSDI, ASPLOS, ICSE, WWW, etc., and won more than 10 honors for outstanding papers, including the WWW Best Paper Award. He has won the first prize of the National Technological Invention Award, the "Young Scientist Award" of the Ministry of Education, the "Young Scientist Award" of the CCF-IEEE CS, the IEEE TCSC Mid-career Excellence Research Award for Scalable Computing, the "Top Ten Teachers" of Peking University, and the "Teaching Excellence Award" of Peking University. He has presided over a number of national, provincial and ministerial scientific research projects; He has served as the Associate Editor of IEEE TMC, Science China Information Science and other journals, and the chair/member of the program committee of several international conferences.
Title: System Software in the Era of Large Models: Thinking and Practice
Report Summary: Large models and their applications are driving a new wave of artificial intelligence. With the increasing complexity of the model structure and the continuous growth of the application scale, from the perspective of system resource management, there are the "three nos" challenges of "not enough", "not easy to use" and "not good to use", and computer software technology is facing a new round of changes and development opportunities. This report will introduce the team's research progress in cross-domain resource hybrid service programming, training/training and push-integrated scheduling optimization based on large-scale (Vanka-level) clusters, and collaborative scheduling of multi-model complex tasks from the perspective of the large model system software stack.
Hu Chunming
Beijing University of Aeronautics and Astronautics
Hu Chunming, Director of CCF, Deputy Director of CCF System Software Committee, Professor and Doctoral Supervisor of Beihang University. Deputy Director of Beijing Advanced Innovation Center for Big Data Science and Brain-Computer Intelligence. Director of the Working Committee for the Popularization of Computer Science, Deputy Secretary-General of the Cloud Computing and Big Data Expert Committee of the Chinese Institute of Electronics, and Vice Chairman of the International World Wide Web Consortium (W3C). His research interests include computer software and theory, distributed systems, virtualization of computing systems, data center resource management and scheduling, graph computing query optimization, etc. He has presided over 6 national projects (and topics), presided over 1 key project in Beijing, and obtained 35 authorized technical invention patents. The research on network resource organization, management and scheduling technology won 1 second prize of the National Technological Invention Award (the 2nd completer), the research on the hybrid scheduling technology of large-scale data center in offline load won 1 special prize of the Chinese Institute of Electronics Science and Technology Progress Award (the 3rd completer), and the research on behavioral big data fusion processing technology won the first prize of the Technical Invention Award of the Chinese Institute of Electronics (the 3rd completer). Selected for the 14th China Youth Science and Technology Award.
Title: TBD
Report Summary: TBD
Zhai Ennan
Alibaba Cloud
Zhai Ennan is a member of the Standing Committee of the CCF Internet Committee and the head of Alibaba Cloud Network Research. He received his Ph.D. in Computer Science from Yale University in 2015 and subsequently served as a research assistant professor at Yale University. He joined Alibaba in 2018 and led Alibaba Cloud's intelligent network reliability O&M system to ensure zero network failures during the Beijing Olympics, Tokyo Olympics, and online office for hundreds of millions of people. His research interests include computer networks, distributed systems, and program verification, and he has published more than 60 papers in top international conferences in these fields, such as SIGCOMM, OSDI, SOSP, NSDI, etc. He has served as a member of the program committee of top international conferences such as SIGCOMM and NSDI for many times. He is currently a member of the Standing Committee of the CCF Internet Committee. He won the first prize of technological invention of the Communication Society once.
Title: Heterogeneous Computing Abstraction System for Computing Power Network
Abstract: With the rapid development of artificial intelligence and cloud computing, providing high-performance computing power and scheduling and then establishing computing power networks has become one of the important trends in the development of computer science in mainland China. The construction of the computing power network requires the design and development of multi-level and multi-dimensional system abstraction, so as to shield the details related to the underlying computing power (including heterogeneous chips, remote nodes, etc.) and empower the upper-layer computing power management. This report starts with a macro computing network abstraction system scheme, and then delves into the abstraction system design for heterogeneous computing chips (including FPGAs, ASICs, and DPUs), which includes advanced technologies such as cross-platform heterogeneous chip programming, correctness verification testing, and program orchestration and deployment (all of which have been published in SIGCOMM and NSDI, the top conferences of network systems in the past few years). Finally, a possible open question is given for a direction that Alibaba Cloud is researching, "Collective Communication Optimization Generator for Heterogeneous Clusters".
Situation Wenchuan
CLP Jinxin Research Institute
Guo Wenchuan, Chief Scientist of China Electronics Co., Ltd., President of CLP Jinxin Research Institute. He has been engaged in financial IT services for 32 years, successively served as the general manager, vice president, general manager and CTO of the R&D center of two listed companies, and co-founded a financial IT service company. He is also the vice chairman of the Standard Professional Committee of Zhongguancun Fintech Industry Development Alliance, the executive member of the CCF Distributed Computing and System Committee, and the executive member of the Open Source Development Committee. He has successively engaged in the research and development of network communications, core banking systems, counter systems, middleware architecture design, CRM, wealth management, mobile finance and other systems, as well as business consulting and IT planning services.
Title: Practical Exploration of Hash Network Technology in the Financial Industry
Report Abstract: At present, the demand for computing power is growing exponentially, the interconnection of computing power has become a common trend, and all walks of life are actively exploring the construction of digital information infrastructure under the transformation and upgrading of computing power. When the financial industry uses technologies such as computing power network to cope with the diversified evolution of computing power elements and computing architectures, it is faced with many industry characteristic requirements and practical challenges, and it needs a systematic approach that integrates the concepts of system engineering and platform engineering to achieve joint technical research, verification of key scenarios, sharing of computing power resources and replication of industry capabilities, and building a new digital infrastructure for the intelligent era.
Chen Quan
Shanghai Jiao Tong University
Chen Quan, Ph.D., is a professor and deputy director of the Department of Computer Science, Shanghai Jiao Tong University. He has been engaged in research related to computer architecture and cloud native computing for a long time. He has presided over the key projects of the National Natural Science Foundation of China, and published more than 100 academic papers in famous international conferences and journals in the fields of ASPLOS, OSDI, ATC, SC, TC, TPDS, TACO, etc. He has won the CCF Youth Science and Technology Award, the Ali Green Orange Award, and the IEEE TCSC Young Scholar Award. At present, he is a member of the youth editorial board of Fundamental Research, a journal sponsored by the National Natural Science Foundation of China, and an editorial board member of the SCI journals Parallel Computing, JCST, FCS and youth editorial board members in the field. His research achievements have won the first prize of the 2023 CCF Technological Invention Award (ranked 1st), the second prize of the National Technological Invention Award, and a number of provincial and ministerial science and technology first prizes.
Title: Scheduling and Parallel Co-design for Large-Scale Model Training
Report Summary: With the iteration of hardware and interconnect topologies, the heterogeneity of GPU clusters is becoming increasingly significant. On heterogeneous clusters, considering the multiple parallel paradigms of large-scale models, adaptive parallelism can significantly improve the training efficiency of large-scale models. However, the heterogeneity of clusters makes the adaptive parallel search space grow explosively, and it is difficult to determine the optimal parallel mode in a reasonable time. To this end, this report will introduce a heterogeneous cluster scheduling method based on scheduling and adaptive parallel co-optimization. By redefining the scheduling unit of cluster scheduling, combined with the efficient performance modeling method based on execution flow and the elastic scheduling strategy, the training throughput of heterogeneous clusters can be significantly improved.
How to register for the conference
1. Registration Criteria
| Ticket type: | Participant Identity | 6.1~7.28 |
| Conference registration fee | Professional Member of CCF | ￥2700 |
| CCF Student Member | ￥1700 | |
| Non-Member Professionals | ￥3200 | |
| Non-member students | ￥2200 |
2. How to register
| Conference registration QR code | QR code for the homepage of the conference |
Note: The conference is paid through the CCF conference management system, and the refund shall comply with the "CCF Regulations on the Refund of Conference Registration Fees".
For the specific arrangements of the above forums, please pay attention to the official website of the conference. In addition to the keynote speeches and this forum, CCF Computility 2024 will also organize 14 unique thematic forums, each of which will be chaired by top experts to bring you the most cutting-edge academic discussions and technical exchanges.
At present, the preparatory work of the conference is nearing the end, and scholars in the field of distributed computing are welcome to come to Changchun to witness and promote technological innovation.