Three NUS researchers’ research work were displayed at the exhibition:

1. Flow Interference and Thruster-Hull Interactions
   Dr. Zhang Qin and Dr. Rajeev Kumar Jaiman
   Department of Mechanical Engineering and Keppel-NUS Corp Lab, NUS
2. Cross-talking of RNA Editing and Alternative Splicing in Human Tissues
   Kar-Tong Tan, Omer An, Yang Liu, Ying Li, Jia Li, Chan-Shuo Wu, Siqin Zhou, Tony Tan, Henry Yang (Cancer Science Institute of Singapore, NUS)
3. Structure-based Virtual Screening by Distributed Docking Calculations on NCCS HPC
   Po-Hsien Lee, Ley Moy Ng and Bin Tean Teh
   (Cancer Science Institute of Singapore, NUS; Laboratory of Cancer Epigenome, National Cancer Centre Singapore; Division of Cancer and Stem Cell Biology, Duke-NUS Graduate Medical School; Institute of Molecular and Cellular Biology)

The three projects show how they’ve benefited from the NSCC large scale computing system and the petabyte scale of storage.  For example, the first project is to do simulation on a dynamic positioning system which maintains a semi-submersible vessel’s position in the ocean condition by using multiple thrusters. This project investigates the complex flow interactions between thruster-thruster and thruster-hull.  With the help of NSCC supercomputer, the team was able to numerically simulate the full geometry thruster with dynamic mesh by using thousands of CPU cores through paralleling computing and simulation. Without the NSCC supercomputer system, the simulation of such a system would take ages to complete, as the simulation for a single thruster already need hundreds of CPU cores running for a few days.

The projects spark the interest of researchers as well as resource providers in the conference, and attracted people from different parts of the world.

During the six days conference period, some booths were always crowded with visitors. One of the most popular booth was the booth setup by Nvidia, the designer and producer of GPU (Graphics Processing Units) for Computer Game and graphics processing.

With the exciting progress of deep learning and other machine learning algorithms, the artificial intelligence (AI) started to play a very important part in people’s life, from doing simple tasks such as image classification, video analytics, speech recognition and natural language processing to more advanced applications like self-driving car and board games.

Over the past few years AI has exploded, especially since 2015. Much of that has to do with the wide availability of parallel processing GPUs and the flood of big data. Amazon, Baidu, Google, Facebook and Microsoft were the first adopters of NVIDIA GPUs for deep learning. With the opening of their deep-learning platforms for all to use, AI-powered applications will spread fast. I believe this is why Nvidia booth attract more people; as more and more people are involved in the AI development now.

Other popular booth included DDN, the provider of high performance, high capacity big data storage systems, and Amazon, which demonstrate their real-world examples of HPC in the Cloud.

The announcement of the latest edition of the TOP500 is one of the few activities that attracted most people’s attention during Supercomputing Conference. The 48th edition of the TOP500 list at SC16 saw China and United States pacing each other for supercomputing supremacy. Both nations claim 171 systems apiece in the latest rankings, accounting for two-thirds of the list. However, China has maintained its dominance at the top of the list with the same No. 1 and 2 systems from six months ago: Sunway TaihuLight, at 93 petaflops, and Tianhe-2, at 34 petaflops. This is in comparison with the list published in November 2008, when the top ten were all from the US institutes and China only had 15 systems on the list. In November 2016, it reached parity with the US on the TOP500 list in total number of systems and aggregate performance. It can be seen how China has leapfrogged the HPC competition in the last ten years.

I am glad to see that the NSCC supercomputer manage to grab the 115^th position, with 30,912 processor cores and 1.01 petaflops, quadrillions of calculations per second, on the Linpack benchmark in the 48^th TOP500 list. This very powerful supercomputer is now available to all the NUS researchers for running large scale computing intensive and memory intensive calculations, and terabytes of data processing.

The competition was a real-time, non-stop, 48-hour challenge where teams of six undergraduates assembled a small cluster on the exhibit floor and raced to complete a real-world workload across a series of scientific applications, demonstrated knowledge of system architecture and application performance, and impressed HPC industry judges.

Colleague kids, team work, caffeine, no sleep, challenges and more – SC16 was the 10^th year that the Students Cluster Competition has been running in the conference. This year, 14 student teams from different parts of the world came to compete for the top prizes in two categories: Best Performance of the LINPACK benchmark application and Best Overall team performance. The University of Science and Technology of China took the top honor in both categories this year.

I hope NUS students who are passionate with the high performance computing can participate in the competition at the SC17 which will be held at Denver in November 2017.