Jack Dongarra
Depei Qian
Marek Michalewicz
Hai Jin
Yutong Lu
Zeyao Mo
José A.B. Fortes

ASC16 Final Competition

April 18-22, 2016

International Academic Exchange Center Building 8, Academic Lecture Hall on the 3rd Floor, Huazhong University of Science and Technology

The ASC Student Supercomputer Challenge is the world’s largest supercomputing hackathon. This year, the 16 finalists are selected among all 175 registered teams worldwide after two months of rigorous evaluation in the preliminary contest. During the final round, the teams will design and build a cluster to conduct a benchmark test HPL, a benchmark standard HPCG, a surface wave numerical model MASNUM, a material simulation software ABINIT, a deep learning algorithm DNN, and a Mystery Application under 3000W. Inspur Group will provide their server and other necessary equipment. At the end, all teams will be required to present to the Judge Committee and introduce their system designs, the final result and analysis.


We welcome you to attend the ASC16 Awards Ceremony on April 22 and join the On-Site Visit anytime from April 18 to April 22. RSVP

The 12th HPC Connection Workshop

HPC: Challenges & Opportunities

April 21, 2016

Huazhong University of Science and Technology, Wuhan, China

The 12th HPC Connection Workshop, an international High-Performance Computing summit, is organized jointly by the Asia Supercomputer Community and Inspur Group. The event takes place three times a year: during the International Supercomputing Conference (ISC), the Supercomputing Conference (SC), and the ASC Challenge.

14:00 Qingming Luo
Vice President, Huazhong University of Science and Technology
Welcome Speech
14:10 Jack Dongarra
University of Tennessee
Oak Ridge National Laboratory

Current High-Performance Computing Trends and Challenges for the FutureDownload

In this talk we examine how high performance computing has changed over the last 10-year and look toward the future in terms of trends. These changes have had and will continue to have a major impact on our numerical scientific software. A new generation of software libraries and algorithms are needed for the effective and reliable use of (wide area) dynamic, distributed and parallel environments.  Some of the software and algorithm challenges have already been encountered, such as management of communication and memory hierarchies through a combination of compile--time and run--time techniques, but the increased scale of computation, depth of memory hierarchies, range of latencies, and increased run--time environment variability will make these problems much harder.
14:50 Depei Qian
Professor, Beihang University, Sun Yat-sen University
Director of the Key Project on HPC, National High-Tech R&D Program

China HPC Development in the 13th 5-Year PlanDownload

After a brief review on HPC research and development in China’s high-tech R&D program, this talk will introduce the plan of HPC development under the new Key R&D Program of China in the 13th 5-year plan. The major challenges and opportunities in establishing the eco-system for high performance computing in China will be discussed, including the technical issues in developing the next generation HPC system using domestic processors, the urgency of developing application software for the domestic processor-based system, and the difficulty in establishing a national HPC environment with sustainable operation model and mechanism. The goal and the major activities of the new key project on HPC will be briefly presented.
15:30 Marek Michalewicz
Chief Executive Officer, A*STAR Computational Resource Centre
InfiniCortex: A Global Supercomputer
A*CRC has built InfiniCortex, a global supercomputer platform that combines very high bandwidth networks, storage and supercomputing resources spread over four continents. This talk will share to create an alternative path to reach Exascale: global scale concurrent supercomputing utilising efficient, lossless and encrypted trans-continental InfiniBand allowing RDMA; and a Galaxy of Supercomputers (a topological abstraction). Geographically distributed HPC and Storage resources are connected into a single concurrent and distributed compute resource, alleviating the pressures of colossal power as well as infrastructure, data replication and disaster recovery requirements. The talk will report on recent breakthroughs of InfiniCortex: i) ring-around-the-world connectivity using exclusively InfiniBand transport with trans-Pacific and trans-USA links at 100Gbps; ii) the first InfiniBand routing with seven separate sub-nets; iii) InfiniCloud: HPC cloud underpinned with InfiniBand fabric instances with no limit on size of resources and spanning four continents; iv) Rich application layer run on the InfiniCortex infrastructure.
16:00 Hai Jin
Cheung Kung Scholars Chair Professor, Huazhong University of Science and Technology
Architecture Consideration for Big Data Processing
With emerging of big data, the processing speed for the data is one of the key issues for big data technology. One of the efficient ways to handle the velocity of data is putting all the data in the memory. But traditional memory, DRAM, consumes a large amount of energy and cost to build a large memory system. In recent years, lots of non-volatile memory devices, such as phase change memory (PCM), are studied to be part of memory. We call them storage class memory (SCM). Combing traditional memory and SCM together to build a large hybrid memory space is becoming one energy-efficient way to extend the traditional in-memory computing system into a new level, to handle large quality of data in real time. In this talk, we will discuss this new in-memory computing system from different aspects and some challenges in this new system. We will also report some ongoing effort in China to build this hybrid memory-based in-memory computing system, and some latest advances in this area.
16:30 Yutong Lu
Professor, Deputy Director of the Software Research Institute, National University of Defense Technology
Deputy Chief Designer, Tianhe Project
Director, National Supercomputing Center in Guangzhou
Convergence of HPC and Big Data
Nowadays, advanced computing and visualizing tools are facilitating scientists and engineers to perform virtual experiments and analyze large-scale datasets. Computing-driven and Big Data-driven scientific discovery has become a necessary approach in global environment, life science, nanomaterials, high energy physics and other fields. Furthermore, the fast increasing computing requirements from economic and social development also call for the birth of the Exascale system. This talk will discuss the convergence of the HPC and Big Data on Tianhe-2 system.
17:00 Zeyao Mo
Vice President, Institute of Applied Physics and Computational Mathematics
Parallel Algorithm and Parallel Programming: A Software Perspective
This talk reports on a software perspective on the evolution of parallel algorithm and parallel programming for numerical simulation. In the past of terascale computing, they are two essential modules of parallel application software. Nowadays, they are encapsulated into a middleware named by the parallel programming framework on which a wide variety of parallel application software can be developed. In the future of exascale computing, they should be built on a run time system for better allocation of computing resources. Here, some experience results are also given.
17:30 José A.B. Fortes
Professor and AT&T Eminent Scholar, University of Florida
Director of the Advanced Computing and Information Systems Laboratory, University of Florida
Concepts and Challenges of Software-Defined Cloud Systems
"Software-definition" is an emerging transformative aspect of IT. It has had a major impact on networking, where major equipment manufacturers already provide products (e.g. routers, switches and controllers) that support software-defined networking. However, even in the context of networking, software-definition is at its infancy, with much research and development being necessary to understand the new capabilities, challenges and applications that it enables in IT systems. In the future we will see a generalization of software-definition to all components of IT infrastructures, software systems and CI applications. Datacenters are already adopting management techniques using software-defined compute, storage and networking – leading to software-defined datacenters and impacting the design and use of IT clouds. More importantly, software-definition is becoming a new dimension of systems design entailing (1) a refactoring of systems that leads to separation of their data and control planes (or, somewhat equivalently, their functional and management aspects), and (2) the exposure of control or management capabilities to users/consumers of the systems. Software-definition introduces major opportunities for research in computer science and engineering, with potential for long-lasting broad impact. This talk will review and exemplify the motivation and nature of software-defined IT systems, and introduce some of the opportunities, research challenges and approaches in the design and management of such systems.

Expenses: The Workshop is free if you sign up before March 31. No other expenses will be covered due to high attendance this year.

Visa: We can issue a letter of invitation if you need to apply for a visa. Please send your information (Name (as in passport), Gender, Date of Birth, Nationality, Passport No., Organization) to executive.director@asc-events.org.


First Name *  
Last Name *  
Nationality *  
Email *  
Job Title *  
Organization *  
Occupation *
I would like to attend the HPC Connection Workshop on Apr 21
the ASC16 Awards Ceremony on Apr 22
* Required


Contact Us

Tech Inquiry
Wenjing Lv & Weiwei Wang

Hybrid Comp Inquiry
Qing Zhang

Media Inquiry
Jie He

Cooperation Inquiry
Xiaochen Fan

General Inquiry
Yan Yan

Other Inquiry
Ying Zhang

Follow Us
Copyright 2016 Asia Supercomputer Community. All Rights Reserved