China Dominates the World TOP500 Supercomputers

[JSCh]

Chinese Company Sugon Placed on US ‘Entity List’ After Strong Showing at International Supercomputing Conference
By Tiffany Trader
June 26, 2019

After more than six years of proclaiming its supercomputing prowess, and operating the world’s most powerful supercomputer from June 2013 to June 2018, China is keeping mum on its latest HPC accomplishment.

According to multiple sources (with further confirmation yesterday from the South China Morning Post), Chinese vendor Sugon (known also as Dawning Information Industry Co.) was poised to stand up the new no. 1 supercomputer at the International Supercomputing Conference in Germany last week. But Sugon pulled back the entry despite attaining benchmarking results 2x higher than the U.S.-based Summit supercomputer, currently ranked no. 1 in the world.

With trade and tech war tensions heightened in advance of U.S. President Donald Trump’s and Chinese leader Xi Jinping’s planned meeting at the G20 summit this month, sources told us that China did not wish to invite scrutiny and potentially face the same sanctions as Huawei. Nevertheless, on Friday (June 21) the day after ISC wrapped up, Sugon and four other Chinese organizations were placed on the U.S. “entity list” by the U.S. Commerce Department, which cited risks to American national security or foreign policy interests. The action effectively bars the named entities from accessing advanced U.S. computing technologies.

​

Sugon demos Nebula supercomputer at ISC 2019

Despite holding back its benchmarking news, Sugon generated a lot of attention at ISC with the demonstration of Nebula, its “Next Generation Silicon Cube High Performance Computer” that features phase change immersive cooling. Standing roughly 4 meters high by 2.5 meters wide, the two-rack system in Sugon’s booth, when fully equipped, holds 160 dual-node blades for a total of 320 nodes. Each node houses one China-made AMD x86 CPU and four AMD GPUs. The AMD CPU is China’s version of the Epyc processor, licensed through AMD’s joint venture with THATIC (a Chinese holding company also added to the entity list). Sugon told HPCwire the system also supports Intel-plus-AMD GPU gear, and booth signage indicated Intel partnership status. The Silicon Cube computer supports fat tree topology and 6D-Torus high-speed network technology with 200 Gbps interconnection, according to Sugon.


...
Chinese Company Sugon Placed on US 'Entity List' After Strong Showing at International Supercomputing Conference | HPCwire

 

[JSCh]

Supercomputing centers unveil new engine for innovation in China
Source: Xinhua| 2019-07-07 22:02:57|Editor: ZX

BEIJING, July 7 (Xinhua) -- China has built six National Supercomputing Centers (NSCC) since 2009, serving as a new driver for the country's innovation, according to the NSCC in north China's Tianjin Municipality, which celebrated the 10th anniversary of the founding of the center on Saturday.

Since the establishment of the NSCC in Tianjin was approved by the Ministry of Science and Technology in May 2009, other five supercomputing centers were founded one after another in Shenzhen, Jinan, Changsha, Guangzhou and Wuxi respectively.

As the first supercomputing center in China, the NSCC in Tianjin is not only where China's first petaflop supercomputer the Tianhe-1 is located, but also responsible for developing China's new generation of the exascale supercomputer the Tianhe-3.

Tianjin has established a complete autonomous information industry including high-performance chips, autonomous control system, high-performance server and database, setting up a model on the transformation of technologic innovation achievements, said Li Xiang, vice president of the National University of Defense Technology.

"The supercomputer has become a symbol of power reflecting the innovative capabilities of China. Next, we will connect these supercomputing centers and share the resources nationwide," said Mei Jianping, deputy director-general of the Department of High and New Technology of the Ministry of Science and Technology.

 

[JSCh]

Supercomputing efforts set to gain momentum in China
By Yang Cheng in Tianjin | China Daily | Updated: 2019-07-17 09:57

Scientists check the condition of the Tianhe-3 supercomputer at the National Supercomputer Center in Tianjin in April. [Photo/Xinhua]

Work on next generation of exascale models likely to be completed by 2021

Research and development work on the next generation of exascale supercomputers, the Tianhe-3, is expected to be completed by around 2021, according to the National Supercomputer Center in Tianjin.

Making the announcement at a ceremony to mark the country's 40-year development in supercomputing, officials said that the new machines will have a calculation speed that is 200 times faster than the present models and have storage capacity that is 100 times more than the Tianhe-1. The speed is expected to hit a billion billions, or 1 followed by 18 zeros.

Insiders indicated that 2021 could see tough competition among the exascale supercomputers around the world in their final stage of research.

The US' first exascale supercomputer will be delivered by 2021, while the European Union and Japan have announced their completion by around 2020.

Liu Guangming, director of the National Supercomputer Center in Tianjin, said the Tianhe-3 has seen several major self-innovations including three chips for many core processors, interconnection and routers; nodes for servers, computing and storage, some 10 printed circuit boards; new hardware systems for computing, high-speed interconnection, parallel storage, service processing, surveillance and analysis; and software systems for system operation, parallel research and development, application support and comprehensive management.

The supercomputer will see widespread applications on the convergence front in cloud computing and big data, computing and stimulation on highly efficient weather forecasts, massive measurements in wind tunnels, earthquake monitoring and geology research, medical care, information security services, intelligent urban services, and effectively advancing the next generation of artificial intelligence development for the country.

The Tianhe-3 based in the Binhai New Area is expected to herald a new chapter for the intelligent manufacturing industry in the city and drive technology innovation among Beijing, Tianjin and Hebei provinces.

Liu said: "The past 10 years have seen an innovation convergence in homegrown computing technologies, high-efficiency semiconductors, high-speed interconnections and operational systems. The next decade will be a golden one for companies like Sugon and Insper, especially for their high speed servers and computing systems."

After Tianhe released the prototype of China's first new-generation exascale supercomputer Tianhe-3 in July last year, the prototypes of other exascale supercomputers of Sunway and Sugon were announced.

Currently, how big data, artificial intelligence and exascale supercomputers can promote traditional industries' transformation and regional development will be the key challenge for the sector, said experts.

Zhang Yunquan, director of the National Supercomputer Center in Jinan, East China's Shandong province, said that China's supercomputers are feeling the pinch from the gaps in applications and software development.

"In areas such as manufacturing industries and basic research sectors where supercomputers are used widely, foreign software companies have dominated the market ... and China needs to make huge investments on software procurement and upgrading.

"In addition, China has a shortage of software development research personnel for supercomputers," said Zhang, adding that China's supercomputer makers should transfer their development model to be more market and application driven.

In addition, focus needs to be paid to software development and education and stimulation of human resources.

Mei Jianping, deputy director-general of the High and New Technology Department of the Ministry of Science and Technology, said the country will continue to support the establishment of the supercomputer centers in key areas, and enhance their collaboration with other key supercomputer centers in China.

To date, China has built six national supercomputing centers since 2009, mainly as new drivers of innovation.

Since the establishment of the center in Tianjin was approved by the Ministry of Science and Technology in May 2009, five other supercomputing centers were founded in Shenzhen, Jinan, Changsha, Guangzhou and Wuxi.

 

[JSCh]

Paving the way for China exascale computing | SpringerLink

 

[JSCh]

China Extends Lead in Number of TOP500 Supercomputers, US Holds on to Performance Advantage | TOP500 Supercomputer Sites
TOP500 News Team | November 18, 2019 08:04 CET

BERKELEY, Calif.; FRANKFURT, Germany; and KNOXVILLE, Tenn.— The 54th edition of the TOP500 saw China and the US maintaining their dominance of the list, albeit in different categories. Meanwhile, the aggregate performance of the 500 systems, based on the High Performance Linpack (HPL) benchmark, continues to rise and now sits at 1.65 exaflops. The entry level to the list has risen to 1.14 petaflops, up from 1.02 petaflops in the previous list in June 2019.

Top-ranked systems

The top of the list remains largely unchanged. In fact, the top 10 systems are unchanged from the previous list.

Summit and Sierra remain in the top two spots. Both are IBM-built supercomputers employing Power9 CPUs and NVIDIA Tesla V100 GPUs. Oak Ridge National Laboratory’s Summit system holds top honors with an HPL result of 148.6 petaflops. The second-ranked Sierra system at Lawrence Livermore National Laboratory comes in at 94.6 petaflops.

Close behind at number three is the Sunway TaihuLight supercomputer, with an HPL mark of 93.0 petaflops. TaihuLight was developed by China’s National Research Center of Parallel Computer Engineering & Technology (NRCPC) and is installed at the National Supercomputing Center in Wuxi. It is powered exclusively by Sunway’s SW26010 processors.

Tianhe-2A (Milky Way-2A), a system developed by China’s National University of Defense Technology (NUDT) and deployed at the National Supercomputer Center in Guangzhou, China, holds the number four spot with 61.4 petaflops. It is powered by Intel Xeon CPUs and Matrix-2000 accelerators.

Frontera, a Dell C6420 system, keeps it number five ranking, with an HPL result of 23.5 petaflops. Powered solely by Xeon Platinum processors, it was installed at the Texas Advanced Computing Center of the University of Texas in 2018.

At number six is Piz Daint, a Cray XC50 supercomputer installed at the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland. At 21.2 petaflops, it remains the most powerful system in Europe.

Trinity, a Cray XC40 system operated by Los Alamos National Laboratory and Sandia National Laboratories is ranked number seven, with an HPL performance of 20.2 petaflops. It is powered by Intel Xeon and Xeon Phi processors.

Listed at number eight is the AI Bridging Cloud Infrastructure (ABCI), installed at Japan’s National Institute of Advanced Industrial Science and Technology (AIST). The Fujitsu-built system is equipped with Intel Xeon Gold processors and NVIDIA Tesla V100 GPUs, achieving an HPL result of 19.9 petaflops.

SuperMUC-NG is in the number nine position with 19.5 petaflops. It’s installed at the Leibniz-Rechenzentrum (Leibniz Supercomputing Centre) in Garching, near Munich. The Lenovo-built machine is powered by Intel Platinum Xeon processors.

The Lassen supercomputer is ranked number 10, delivering 18.2 petaflops. Installed at Lawrence Livermore National Laboratory, Lassen is the unclassified counterpart to the classified Sierra system and shares the same IBM Power9/NVIDIA V100 GPU architecture.

The most powerful new supercomputer on the list is AiMOS, which shows up at position 24, turning in an HPL result of 8.0 petaflops. The IBM-built system is installed at the Rensselaer Polytechnic Institute Center for Computational Innovations (CCI), and like Summit, Sierra, and Lassen, it is equipped with Power9 CPUs and NVIDIA V100 GPUs.

Geographical breakdown

The number of TOP500 installations in China continues to rise and now sits at 227, up from 219 six months ago. Meanwhile, the share of US-based system remains near its all-time low at 118. However, systems in the US are on average significantly larger, which translated to a 37.8 percent share of the list’s aggregate performance. China is close behind with a 31.9 percent performance share. However, compared to six months ago, this performance gap has shrunk. The June 2019 list had the US with a 38.4 percent of the list’s aggregate performance and China with 29.9 percent.

Japan remains in third place in the number of TOP500 systems, with 29, followed by France with 18, Germany with 16, the Netherlands with 15, Ireland with 14, and the United Kingdom with 11. All other countries were in the single digits.

Vendor shares

As a reflection of China’s dominance in sheer numbers, the top three system vendors with regard to the number of installations are Lenovo, (174), Sugon (71), and Inspur (65). Cray is number four, with 36 systems, and HPE is number five, with 35. Note that Cray is now part of HPE, so taken together they would effectively tie Sugon with 71 systems.

At the chip level, Intel continues its dominance. Its processors are present in 470 of the 500 systems, split between multiple generations of Xeon and Xeon Phi hardware. IBM is second with 14 systems – 10 with Power CPUs and four with Blue Gene/PowerPC CPUs. AMD claims just three systems on the current list.

There are now two Arm-based supercomputers on the list: the Astra system deployed at Sandia National Laboratories, which is equipped with Marvell’s ThunderX2 processors, and Fujitsu’s A64FX prototype system, a precursor to the Fugaku (Post-K) exascale system destined for RIKEN in 2021. The Fujitsu machine is new to the list and occupies position 159, with an HPL performance of 2.0 petaflops.

NVIDIA is the dominant vendor for accelerators. Its GPUs are present in 136 of the 145 accelerated systems. On the previous list six months ago, there were 134 accelerated systems.

Interconnect breakdown

Ethernet is used in 52 percent (258) of the TOP500 systems, while InfiniBand is the network-of-choice in 28 percent (140) of systems. However, from a performance perspective, those positions are reversed, with InfiniBand-based machines representing 40 percent of the TOP500’s aggregate performance and Ethernet-based machines with 29 percent. Custom interconnects, with just 46 installations, claim 22 percent of the list’s installed performance.

Green500 results

The related Green500 list, which measures energy-efficiency of the world’s top supercomputers, has changed considerably compared to six months ago. Number one on the list is the aforementioned A64FX prototype supercomputer, which delivered 16.9 gigaflops/watt. Just slightly less green at number two system is NA-1, a Zettascaler machine that uses PEZY Computing’s PEZY-SC2 processors and delivers 16.3 gigaflops/watt. It is being readied for a future installation at NA Simulation in Japan.

The third-ranked Green500 system is the new AiMOS system from IBM, followed by two more IBM systems based on the same Power9/NVIDIA V100 design: Satori at 15.6 gigaflops/watt and Summit at 14.7 gigaflops/watt. The remaining top 10 Green500 systems on the list – AI Bridging Cloud Infrastructure, MareNostrum P9 CTE, TSUBAME 3.0, PANGEA III, and Sierra – also use NVIDIA GPU accelerators.

HPCG results

The two top-ranked Summit and Sierra supercomputers on the TOP500, also remain in the top two spots on the list based on the High-Performance Conjugate Gradient (HPCG) benchmark. Summit achieved 2.93 HPCG-petaflops, with Sierra at 1.80 HPCG-petaflops. All the remaining top 10 HPCG entries, delivered less than one HPCG-petaflops. With the exception of the now-decommissioned K-computer, all 10 of these systems carried over from the previous list six months ago.

 

[GS Zhou]

The top of the list remains largely unchanged. In fact, the top 10 systems are unchanged from the previous list.

A bit disappointed to see the top list unchanged.

I'm wondering when China's Exaflop Supercomputer could be ready. I heard we have three prototypes under development in parallel, any news on its latest progress??

 

[JSCh]

A bit disappointed to see the top list unchanged.

I'm wondering when China's Exaflop Supercomputer could be ready. I heard we have three prototypes under development in parallel, any news on its latest progress??

No change. All three exascale prototypes had been handed over to user for evaluation. Tianhe-3 exascale would be operational around 2021.

 

[JSCh]

Sugon’s “Silicon Cube” Won the Innovation Excellence Award | TOP500 Supercomputer Sites
Sponsored Content
Dawning Information Industry Co., Ltd. | November 19, 2019 20:26 CET

On Nov. 20th, Beijing time (Nov. 19th Local Time)，SC19 (The Supercomputing Conference 2019) was held in Denver, Colorado, USA. The theme of the conference is “HPC is now”. During the conference, well-recognized Hyperion Research issued Sugon the Innovation Excellence Award, due to its impressive innovations on “Silicon Cube” series computer. As reported, this is the first time that a Chinese local IT enterprise successfully gains the honor.

As commended by the award committee, Sugon’s new generation of “Silicon Cube” series computer have achieved the world-leading computing density and energy efficiency by adopting the full immersion and phase change liquid cooling technology. Compared with its predecessor, the computer has achieved 10X higher performance of one rack, 10-20X higher computing density, and 4X higher system energy efficiency, and 2X better system energy efficiency.

The goal of the Innovation Excellence Award is not limited to recommend the innovative hardware and solutions, which has been recognized and accepted by users, but also to present the successful application of computing in the field of science and related industry. Demonstrating the return of investment on computing shall help the public to better understand the essential value of computing technology.

This award had been selected and assessed by a steering committee composed of dozens of senior users and experts in the field, specifically includes those experts from California Institute of Technology, Virginia Tech University, Stuttgart University, Ford Motor, general motors, Berkeley National Laboratory, Oak Ridge National Laboratory, etc.

The Innovation Excellence Award has been released for 15 times, and ~100 innovative hardware and applications have been listed. The winners include NVIDIA Tesla VI00 AI Acceleration Card, IBM's micro data center, and the applications for seismic simulation & oil exploration, weather & earth science and so on.

According to related survey, 30 to 50% of the energy consumption of a data center has been utilized for the cooling system in recent years. Compared with the traditional air cooling system, the “Silicon Cube” with the full immersion and phase change liquid cooling technology is capable of saving 30% of the energy consumption of cooling. Meanwhile, the energy efficiency ratio (PUE) can be as green as 1.04 (the average PUE value of the global data center is 1.58).

The full immersion and phase change liquid cooling technology further improves the heat dissipation efficiency of the core components in data center. This advanced technology can increase the frequency of the CPU by 15 to 20%, bring 20% to 40% improvement on application performance, and reduce the cost of chips by 30% to 60%. More stable temperature for the system operating environment can elongate the life of electronic components inside. Noted that the distinct feature of nearly zero noise computer room could provide a much friendly environment for system operators and users.

In order to address the computing issues on energy consumption, sustainability, reliability, and the high costs of system operation & maintenance, Sugon’s “Silicon Cube” successfully examples the promising solutions with liquid cooling technology, hierarchy-module-3D innovative design. During the R&D of “Silicon Cube” computer, Sugon has accumulated large amounts of future-oriented technology patents. For instance, in the field of “liquid cooling”, it hosts >40 core patents.

Leveraging the excellent energy-saving efficiency and innovative “Silicon Cube” computing architecture design, Sugon has already won the awards of “World Leading Internet Scientific and Technological Achievements” and “China's Leading Big Data and Intelligent Computing Enterprise” at the 6th World Internet Conference and the 1st China Super Computing Conference (ChinaSC 2019), respectively.

 

[JSCh]

Chinese media reported that Tianhe's I/O system won 1st in bandwidth and overall 3rd in IO500 2019.

​

Full list here -> https://www.vi4io.org/io500/start

 

[Han Patriot]

They had a faster supercomputer than the US ones but had to postpone it due to the trade war, qe cant appear to beat US again. Haiiih, we are living in such interesting times.

 

[Beast]

They had a faster supercomputer than the US ones but had to postpone it due to the trade war, qe cant appear to beat US again. Haiiih, we are living in such interesting times.

Yes, on paper it seems US is on top but reality is that world fastest super computer is from China in 2019. @GS Zhou

https://www.scmp.com/tech/policy/ar...ot-fan-flames-super-computing-rivalry-amid-us


Chinese decision makers decided to withhold the country’s newest Shuguang supercomputers from the latest supercomputing contest, even though they operate more than 50 per cent faster than the best current US machines, as China does not want to fan the flames of existing trade tensions, said the sources, who declined to be named as the information is private.

 

[Han Patriot]

Yes, on paper it seems US is on top but reality is that world fastest super computer is from China in 2019. @GS Zhou

https://www.scmp.com/tech/policy/ar...ot-fan-flames-super-computing-rivalry-amid-us


Chinese decision makers decided to withhold the country’s newest Shuguang supercomputers from the latest supercomputing contest, even though they operate more than 50 per cent faster than the best current US machines, as China does not want to fan the flames of existing trade tensions, said the sources, who declined to be named as the information is private.

There used to be a time when beating the US sounded so absurd that people would laugh. If 10 years ago you told someone China had too appear weaker on purpose, indians would have diarrhea from laughing too much. Xiaomi becoming an Indian household brand? They would bet their balls against it.... Times have really changed, a mighty US need to use a trade war to contain Chinese power. Damn we are just a poor farming country and we only make plastic toys..... Come on!

 

[Beast]

There used to be a time when beating the US sounded so absurd that people would laugh. If 10 years ago you told someone China had too appear weaker on purpose, indians would have diarrhea from laughing too much. Xiaomi becoming an Indian household brand? They would bet their balls against it.... Times have really changed, a mighty US need to use a trade war to contain Chinese power. Damn we are just a poor farming country and we only make plastic toys..... Come on!

China don't like showmanship. They dont mind others take all the fame on surface but not the real core result in reality. I don't think trade war has any effect on their decision not to participate in that ranking. This no.1 supercomputer is going to contribute a lot of things for China even it will never officially recognized as world fastest in 2019.

 

[GS Zhou]

Yes, on paper it seems US is on top but reality is that world fastest super computer is from China in 2019. @GS Zhou

https://www.scmp.com/tech/policy/ar...ot-fan-flames-super-computing-rivalry-amid-us


Chinese decision makers decided to withhold the country’s newest Shuguang supercomputers from the latest supercomputing contest, even though they operate more than 50 per cent faster than the best current US machines, as China does not want to fan the flames of existing trade tensions, said the sources, who declined to be named as the information is private.

Silence makes big money! 闷声大发财！

 

[Han Patriot]

At least now there will be no more competition for local processors, previously sugon n gang promoted intel n amd.

At least now there will be no more competition for local processors, previously sugon n gang promoted intel n amd.

At least now there will be no more competition for local processors, previously sugon n gang promoted intel n amd.

At least now there will be no more competition for local processors, previously sugon n gang promoted intel n amd.

At least now there will be no more competition for local processors, previously sugon n gang promoted intel n amd.