China Dominates the World TOP500 Supercomputers

[TaiShang]

China is building a 1 billion yuan (US$145.4 million) “superconducting computer” – an unprecedented machine capable of developing new weapons, breaking codes, analysing intelligence and – according to official information and researchers involved in the project – helping stave off surging energy demand.

If successful, that would be great to defend the country's interests and uphold regional and global peace.

Do people in HK have this type of loser mentality?

After 100 years of slumber and miseducation, they are probably slowly coming back to their real historical self.

May take yet another generation, though.

 

[Jlaw]

If successful, that would be great to defend the country's interests and uphold regional and global peace.



After 100 years of slumber and miseducation, they are probably slowly coming back to their real historical self.

May take yet another generation, though.

I'm not surprised. Nothing scientific comes out of HK anymore. Only bank and MLM jobs left.

 

[JSCh]

China Sets Sights on Superconducting Computer | HPCwire
By John Russell
August 28, 2018

China has embarked on a $145 million (~ ¥1 billion) effort to build a superconducting computer with plans for a prototype to be up and running as early as 2022, according a recent article in the South China Morning Post.

“Chinese scientists have already made a number of breakthroughs in applying superconducting technology to computers. They have developed new integrated circuits with superconducting material in labs and tested an industrial process that would enable the production of relatively low cost, sophisticated superconducting chips at mass scale. They have also nearly finished designing the architecture for the computer’s systems,” according to the report, written by Stephen Chen.

Multiple media outlets picked up the article. Chen reports the program was “quietly launched” by the Chinese Academy of Sciences (CAS) in November last year. A superconducting-based computer, of course, would consume far less power and potentially operate at higher frequencies. The power wall is a challenge faced by all supercomputers. Exascale computing projects worldwide are struggling for ways to reduce power consumption. The Department of Energy has set a target of ~40 megawatts for U.S. exascale machines but many believe that is unrealistic. By comparison pre-exascale Summit (U.S.) and Sunway TaihuLight (China) – the top two machines on the current Top500 list – have budgets well under 20MW.

Hyperion Research analyst Bob Sorensen noted, “This is an interesting development. The promise of superconducting components – specifically high-speed and low power – has been around for decades, but high cost and complex technical issue have prevented the development of any serous systems to date. However, the time may be right for this technology to come to the fore due to a confluence of events:

• The slowdown in performance gains in traditional ‘room-temperature’ silicon.
• The growing interest in new exotic technologies and material research to support a ‘Post Moore’s law’ hardware base
• The profusion of superconducting quantum computing efforts that have extended the state of the art in cooling technology that can be applied to superconducting silicon schemes.”

Few details of the China project were presented in the SCMP article. CAS president Bai Chunli is quoted saying the technology could help China challenge the US’ dominance of computers and chips as seen in this excerpt:

“The integrated circuit industry is the core of the information technology industry … that supports economic and social development and safeguards national security,” Bai said in May during a visit to the Shanghai Institute of Microsystem and Information Technology, a major facility for developing superconducting computers…“Superconducting digital circuits and superconducting computers … will help China cut corners and overtake [other countries] in integrated circuit technology,” he was quoted as saying on the institute’s website.

Sorensen also struck a cautionary note, “Although it is too early to tell if this, or any other on-going, superconducting supercomputer research effort will bear fruit in the next five years or so, it is clear that China is looking to expands its advanced computing research efforts across a broad front that includes quantum computing, traditional silicon-based HPCs, and now superconducting systems, and they appear to have a sufficient level of funding to seriously address this multifaceted, technically-demanding, but promising technology for next generation computing.

“Demonstrated success by the Chinese to build a workable superconducting supercomputer ahead of US or other national efforts could give the Chinese a significant lead in the development of high-end HPCs writ large in the next decade. Unlike quantum computers, which require a complete new algorithmic and application paradigm, superconducting supercomputers will be able to use the same wide base of existing algorithms and applications as traditional HPCs, many that support important national security missions.”

The new China machine will indeed rely on “classical structure” and not upon more distant technologies such as quantum computing and neuromorphic computing according Li Xiaowei, executive deputy director of the State Key Laboratory of Computer Architecture, cited in the article.

Chen also reported that Fan Zhongchao, researcher with CAS’s Institute of Semiconductors who reviewed the contract as part of an expert panel, said the hardware was a field-programmable gate array (FPGA), a reconfigurable chip that could be used to simulate and test the design of a large-scale, sophisticated integrated circuit. “The overall design [of the FPGA testing phase] is close to complete,” he is quoted as saying.

If China’s efforts are successful, wrote Chen, the “Chinese military would be able to accelerate research and development for new thermonuclear weapons, stealth jets and next-generation submarines with central processing units running at the frequency of 770 gigahertz or higher. By contrast, the existing fastest commercial processor runs at just 5Ghz.”

Link to South China Morning Post article: https://www.scmp.com/news/china/soc...d-us145-million-superconducting-computer-will

 

[JSCh]

China Aims to Be Global Leader in Superconducting Computers | TOP500 Supercomputer Sites
Michael Feldman | August 28, 2018 18:49 CEST

China is investing $145 million to become a world leader in superconductor-based computing, a technology that could make semiconductor-powered supercomputers and datacenter servers obsolete.

The research project was first reported on August 26 in the South China Morning Post, which noted the effort was “quietly launched by the Chinese Academy of Sciences (CAS) in November last year with a budget estimated to be as much as one billion yuan.” According to the report, Chinese scientists have already developed integrated circuits built from some unnamed superconducting material and have tested an industrial process that would enable low-cost production. Last year, the effort produced computer chips with 10,000 superconducting junctions. The report goes on to say to that a system architecture based on the technology is nearly complete.

Superconducting circuitry is one of a handful of technologies that could maintain the forward progress of computing after Moore’s Law fizzles out in the next decade. Unlike semiconductors, superconductors exhibit almost no electrical resistance, which allows them to operate with extreme energy efficiency. According to a 2017 research paper on the technology, “a superconductor computer could outperform its semiconductor counterparts by two orders of magnitude in energy efficiency, showing 250 GFLOPS/W.” That’s five times the efficiency of a nominal 20 MW exaflop supercomputer at 50 GFLOPS/W, an ambitious metric unlikely to be attained in the first crop of exascale machines that are expected to come online between 2020 and 2023.

The exceptional energy efficiency of superconductors also makes it possible to build extremely fast logic circuits, up to 770 GHz for a particular type of technology known as RSFQ logic, which is based on Josephson junctions. Such speeds would revolutionize supercomputers, which are currently powered by conventional CPUs and GPUs running at 1 to 5 GHz. Processors based on semiconductors haven’t gotten appreciably faster in over ten years, corresponding to the time when Dennard scaling broke down.

Unfortunately, building general-purpose superconducting computers has thus far eluded computer scientists, despite the fact they have been working on technology since the 1950s. For one thing, the materials used to support superconducting requires that it operate at near absolute zero, requiring the computer to be chilled with a cryogenic refrigeration unit. This technology is actually fairly well-developed and, although it adds some complexity and extra power to the system, it’s certainly not a show-stopper.

The larger challenge is the reliance on exotic materials based on niobium, tantalum, and other metallic compounds and using them to manufacture well-behaved superconducting logic circuits and memory cells. A nice summary of superconducting computer research that was eventually abandoned is chronicled in a 2016 IEEE Spectrum article , which traced the various efforts at Bell Telephone Laboratories, IBM, MITI in Japan, and the National Security Agency (NSA).

Interest by the NSA reflects the promise of these of systems for performing intelligence analysis and encryption/decryption at lightening speed. The more recent Chinese interest in superconducting is also being driven by intelligence and security concerns, as well as applications in weapons development. But Li Xiaowei, the executive deputy director of the State Key Laboratory of Computer Architecture, said “the main motivation to build a superconducting computer was to cut the energy demands of future high-performance computers.”

For the Chinese, the research effort also reflects a broader incentive to develop computer technologies that challenge the dominance of US chipmakers. CAS president Bai Chunli noted that superconducting computers will “help China cut corners and overtake [other countries] in integrated circuit technology.”

Nonetheless, the Chinese don’t appear to be overconfident about their chances of outrunning the US and Japan, which have a much longer history in researching and developing superconducting computers. In the US, the current effort to develop this technology is being spearheaded by the Intelligence Advanced Research Projects Activity (IARPA), a federal research agency that fulfills the same role that DARPA does for the Department of Defense.

The IARPA effort, known as Cryogenic Computing Complexity (C3), is primarily focused on developing superconducting logic and memory for high performance computing. Prime contractors include IBM, Northrop Grumman Systems Corporation, and Raytheon BBN Technologies. One of the project’s major accomplishments is the production of a niobium-based digital superconducting chip comprised of more than 70,000 Josephson junctions. The chip was manufactured by Lincoln Laboratory’s niobium foundry, which claims to be the most advanced of its kind in the world.

However, the budget for C3 and even its precise timeline have not been made public. In that respect, the Chinese have been more forthcoming. Although the details of what the Chinese project has produced is still mostly a mystery, the one-billion-yuan effort is planning to unveil a prototype computer as early as 2022. It’s conceivable that the C3 project will also deliver its prototype within the next few years, although, again, IARPA has kept those plans under wraps. The bottom line is that if either of these projects fulfills its promise early in the next decade, the era of exascale computing is going to look a lot different than most people ever envisioned.

Image: Supercomputing circuitry based on Josephson junctions. Source: IARPA

 

[JSCh]

Hyperion Provides Update to Exascale Efforts in US, China, Japan, and Europe
Michael Feldman | September 6, 2018 21:10 CEST

According to the latest analysis from Hyperion Research, the various global efforts to reach exascale supercomputing are making good headway. But in some cases, the decision to develop domestically-produced processors for these systems and the inclusion of new application use cases appears to be stretching out the timelines.

The analysis was presented by Hyperion analyst Bob Sorensen during the company’s latest HPC User Forum event, which was held in Detroit earlier this week. In the presentation, now posted on YouTube, Sorensen provided an update of the exascale development work taking place across the four major supercomputing geographies: the US, China, Japan, and Europe.

According to Sorensen, Hyperion is still projecting that China will produce a “peak” exaflop supercomputer by 2020, with US following in 2021. None of the efforts is expected to produce a sustained exascale machine until 2021, sustained exascale being defined as one exaflop of 64-bit performance on a real application. The slide (below) breaks down the various efforts with regard to projected system deployment dates, processor technologies, system suppliers, and approximate cost per machine.

Source: Hyperion Research​

One thing Sorensen pointed out was that the deployment dates for the European, Japanese, and Chinese systems appeared to be moving out because all three are engaged in significant R&D efforts to build customized HPC processors for their respective systems.

In particular, the deployment dates for the EU machines have pushed out a year or two – to 2023 and 2024 – due to a relatively recent decision to develop domestically produced chips for their first batch of exascale systems. The processor development is being accomplished under the European Processor Initiative (EPI), which, as we reported in July, has just gotten underway. As we noted in that report, the EPI work is expected to result in a general-purpose processor based on the Arm architecture, as well as an accelerator using a RISC-V implementation.

The Chinese exascale effort is focused around three projects, led by the National University of Defense Technology (NUDT), Sunway, and Sugon, respectively. Like the EU systems, these machines will be based on processors designed and manufactured indigenously, most likely based on customized implementations of Arm, x86, and ShenWei architectures. “But we’re starting to see that those programs are slipping,” said Sorensen. “And the early peak exascale ambitions of 2020, I suspect, aren’t going to happen.”

Sorensen speculates that at least part of this delay is related to the new use cases these new Chinese architectures will have to address, namely, the AI/machine learning/deep learning application troika as well as high performance data analytics. As a result, he thinks that it may be 2021 or perhaps even 2022 before China’s first peak exascale system come online, representing a one or two-year delay based on the original schedule.

Japan, meanwhile, is currently on track to put its first exascale supercomputer into production in 2022. That system, known as Post-K, was originally slated to be up and running in 2020, but in 2016, the Japanese had already conceded that they were probably one or two years behind schedule. Sorensen now thinks the 2022 date is rather conservative and they may, in fact, be able to boot up the machine somewhat earlier. Some of that optimism may be related to the fact that the development of the A64FX chip tasked to power the Post-K system is pretty far along, as demonstrated at the recent Hot Chips conference in August.

Ironically, after a somewhat rocky start with questionable government funding, the US exascale plans now look to be on among the most stable. The country’s first peak exascale supercomputer, the A21, is set to deploy at Argonne National Lab in 2021, with additional machines providing sustained exascale performance slated to be installed at other Department of Energy labs in the 2022-2023 timeframe.

Some of this stability can be attributed to the fact that the US has established processor vendors (Intel, NVIDIA, AMD, and IBM), with a long history with high-end chip development, not to mention a choice of multiple system vendors (Cray, IBM, HPE, and Dell) that are able to integrate these chips into cutting-edge HPC machinery. And at this point, the US is already nearly a fifth of the way to exascale with the newly christened Summit supercomputer at Oak Ridge National Lab.

Approximately 92 percent of Summit’s floating point performance is derived from its V100 GPU accelerators, which has already resulted in the world’s first exascale application, albeit at sub-64-bit precision. The Summit application in question is a comparative genomics code that used the innate ability of the GPUs to perform machine learning computations at well above the chip's peak 64-bit FP rating. As Sorensen noted, the suitability of GPUs to greatly accelerate these kinds of lower precision workloads is serendipitously expanding the HPC application landscape and changing how these machines will be used.

“The exascale systems that we are looking at today for 2021 and 2022 may be doing a whole bunch of different classes of applications in a way that we have never really seen before,” explained Sorensen. “We can’t predict what kinds of applications are really going to be driving these systems as they look at sustained performance.”


Hyperion Provides Update to Exascale Efforts in US, China, Japan, and Europe | TOP500 Supercomputer Sites

 

[JSCh]

Chips with full IPR play key role in China’s supercomputer war with US
By Zhao Juecheng Source:Global Times Published: 2018/9/9 19:08:40

​

A Chinese team is using domestically developed technology to build a next-generation computer that will compete with the US, Japan and the European Union for speed records.

Researchers are competing aggressively to develop exascale supercomputers, capable of a billion billion calculations per second.

Journalists from the Global Times recently visited Jinan, East China's Shandong Province, to see the Sunway exascale supercomputer prototype. This computer is not the fastest in China, but is special because it was built entirely with domestic technology. All of the intellectual property used in the computer is owned by China.

Black and embossed with the gold words "Sunway Exascale supercomputer prototype," the cabinet is as tall as a man and stands in a low-temperature but noisy computer room. Its appearance is not distinguished. Only if visitors opened the door could they have a glimpse of the 32 supernodes inside. Each consists of eight multicore processors, the heart of the prototype.

The new prototype of the Sunway supercomputer, in comparison to the previous generation Sunway BlueLight, is one ninth the physical size, and computes three times faster, at 3,000 trillion times a second.

Although it cannot compete with the fastest full-fledged supercomputer TaihuLight, which operates 30 times faster, this prototype has been made using independent research and development.

"There are independent property rights for the processors, Ethernet switching chips and information processing chips," said Zhang Yunquan, director of National Supercomputer Center in Jinan.

"The computing system, high-speed network, and storage management system are all constructed with domestic Sunway devices," Zhang said. Sunway microchips are made by Wuxi-based Jiangnan Computing Lab.

Domestic competition

China started developing exascale supercomputers in 2016 at three different institutes. Exascale computing refers to computing systems capable of at least a billion billion calculations per second. A prototype co-developed by the national supercomputer center in Tianjin and the National University of Defense Technology was completed on July 22. The prototype in Jinan was completed on August 5. The third prototype undertaken by Beijing-based supercomputer maker Sugon is still in progress.

"These prototypes are the first step of China's exascale supercomputing strategy. The three prototype makers will compete against each other. Two will be chosen to create full-version exascale supercomputers," Pan Jingshan, deputy director of the Jinan center, told Global Times. "The real Sunway exascale supercomputer will be released in 2020 according to the plan."

The three competing prototypes use different technology. Sugon uses X86 processors, an approach that Pan says faces significant technical difficulties.

The Tianjin prototype named Tianhe-3 is equipped with an FT 2000+ processor and Matrix 200+ accelerator. The two parts can operate separately or in collaboration.

The Sunway prototype in Jinan adopted a cohesive structure using a Sunway 26010+ processor, four major cores and 256 secondary cores.

"This system is more integrated and energy-efficient, but the challenge lies in parallel programming," Zhang told the Global Times.

"It is estimated by industry insiders that such a major-minor structure might be most suitable for exascale supercomputers," Zhang said.

Too close to call

As the US regained the crown of fastest supercomputer from China this June, the white-hot competition will continue in exascale supercomputers.

Aurora A21, the first US exascale supercomputing system, located in Argonne National Laboratory, is expected to debut in 2021. Two other systems with different structures are under development in Oak Ridge and at the Lawrence Livermore National Laboratory.

Japan based its exascale supercomputer Post-K on the current K Computer, the world speed champion in 2011. However, project leader Yutaka Ishikawa admitted the original delivery time 2020 may be delayed by one to two years, reported HPCwire.

The European Union has been using US technology in their supercomputers, therefore their research is not making substantial progress, Zhang told Global Times.

"China is definitely on the forefront of exascale supercomputers according to our timeframe," Zhang said. "But China and the US are likely to take the lead alternately in the next decade."

In addition to the great amount of money used for research, each supercomputer receives an electric bill over $15 million a year. Considering the cost, how are supercomputers benefiting common people?

"Our strategy is sky-to-ground, taking both national strategy and civilian use into account," Zhang explained.

The supercomputer located in coastal Shandong Province will facilitate China becoming a world maritime power.

The Sunway exascale supercomputer will support ocean and climate prediction, screening of marine drugs, exploration of seabed strategic resources, prediction of marine ecosystem evolution, and intelligent analysis of big sets of marine data.

The supercomputer will also provide a computing platform for fields including health care, advanced manufacturing, aerospace, climate and meteorology.

 

[cirr]

Ingenious computer cooling cuts PKU's energy costs

2018-09-28 16:23:36 chinadaily.com.cn Editor : Li Yan

The cooling system for the supercomputer Weiming-1 uses water at 45 C to cool down the supercomputer. (Photo/Peking University)

An innovative cooling system for a supercomputer has helped one of the country’s most prestigious universities save energy, reported Science and Technology Daily on Thursday.

The cooling system for the supercomputer Weiming-1 uses water at 45 C to cool down the supercomputer, which can save 600,000 yuan ($87,108) a year on electricity bills for Peking University, Yang Yuanqing, CEO and chairman of Chinese computer giant Lenovo, said on Wednesday.

The sum is half of what the university spends each year on cooling the supercomputer, which was developed by Lenovo and has been in use since January to conduct researches.

Water cooling is a popular method to remove heat generated by the components of a supercomputer in operation.

Traditional water cooling solutions use water at over 10 C to cool down a machine, and the temperature goes up to around 20 C after it absorbs heat generated by the machine, said Li Guoqing, deputy CEO of Lenovo.

However, water at such a temperature cannot be reused, he said.

But for the warm water cooling system, the temperature of the cooling water increases from 45 C to over 50 C after absorbing heat from the supercomputer, so it can be recycled in heating devices or as shower water in buildings.

The power usage effectiveness, or PUE, of the supercomputing center in Peking University is 1.1, which means for every 100 kilowatt-hour of electricity used in computing, 10 kWh will be used for cooling and other purposes, higher than the rate of 2 for traditional air cooling systems, Yang said.

"I hope we can reduce it further to 1.05," Yang added.

http://www.ecns.cn/news/sci-tech/2018-09-28/detail-ifyyknzp7231605.shtml

 

[JSCh]

2018中国高性能计算机性能TOP100榜单揭晓—新闻—科学网
2018 China's high-performance computer TOP100 list announced - News - SciNet
2018中国高性能计算机性能TOP100榜单揭晓

10月18日，2018年全国高性能计算学术年会（HPC China 2018）在山东青岛举行。作为HPC China 2018的重头戏，当天下午，备受业界关注的2018年中国高性能计算机TOP100排行榜揭晓。
On October 18, the 2018 National High Performance Computing Academic Conference (HPC China 2018) was held in Qingdao, Shandong. As one of the industry's focus, the 2018 China high-performance computer TOP100 list was unveiled on that afternoon.

本次中国超算TOP100榜单最大的亮点是，国家“十三五”高性能计算专项课题3个E级超算的原型机系统——神威E级原型机、“天河三号”E级原型机、曙光E级原型机均进入性能排行榜前十，分列第四、第六和第九位。该榜单发布人、中科院计算所研究员张云泉告诉记者，根据历史数据拟合推算，E级超级计算机将可能“在2019年左右出现”。
The biggest highlight of this year China HPC TOP100 list is that all three E-class supercomputer prototypes of the national “13th Five-Year” high-performance computing project—the Shenwei E-class prototype, the “Tianhe No.3” E-class prototype, and the Sugon E-class prototype all entered into the top 10 rankings. Sitting at number fourth, sixth and ninth respectively. The list publisher, Zhang Yunquan, a researcher at the Institute of Computing Technology of the Chinese Academy of Sciences, told reporters that based on historical data projection, E-class supercomputers may "appear around 2019."

榜单的前三名毫无变化，依然分别是部署在国家超级计算无锡中心的“神威·太湖之光”、部署在国家超级计算广州中心的“天河二号”、部署在国家超级计算天津中心的“天河一号A”。
The top three of the list have not changed. They are still the three deployed in the National Supercomputing Wuxi Center, "Shenwei·Taihu Light", deployed in the National Supercomputing Guangzhou Center, "Tianhe No. 2", deployed in the National Supercomputing Tianjin Center. "Tianhe No. 1 A".

厂商份额方面，中科曙光、联想分别以40台超算系统入围并列第一，这也是“中科系双雄”第四次并列榜首，其中曙光更是第9次蝉联该榜单桂冠。去年排名第一的浪潮集团，以12台入围屈居曙光、联想其后，“国产三强”占据整个榜单份额的92%。
In terms of vendor share, CAS Sugon and Lenovo each tied for the first place with 40 supercomputer systems, which is also the fourth time for the “CAS twin” to be tied for first. Among them, Sugon is the ninth time to top the list. Last year's ranked No. 1 Inspur group, has 12 sets of finalists. This "domestic trio" accounted for 92% of the entire list.

值得一提的是，本次发布的榜单中，100%为国产高性能计算机系统，国外厂商无一台入围。张云泉表示，这是中国超算 TOP100榜单第一次实现“全国产”。
It is worth mentioning that 100% of the supercomputers on the list are domestic system, no foreign manufacturers are shortlisted. Zhang Yunquan said that this is the first time that China's supercomputer TOP100 list has achieved “all domestic production”.

从应用领域来看，“大数据/机器学习”仍是当下超算的应用热点，TOP100的超算系统中共有27台系统用于大数据分析与机器学习；用于科学计算的系统今年强势回归，数量由去年占比仅11.3%上升至今年的14%。（赵广立）
On the application field, “big data/machine learning” is still the hotspot of current super computing. There are 27 systems in the TOP100 HPC system for big data analysis and machine learning; the system for scientific computing returns strongly this year. The number increased from only 11.3% last year to 14% this year. (Zhao Guangli)

 

[JSCh]

China launches third prototype exascale computer
Source: Xinhua| 2018-10-22 19:29:28|Editor: Yurou


TIANJIN, Oct. 22 (Xinhua) -- China has launched a third prototype exascale computing machine, the next-generation supercomputer, according to the developer.

The Shuguang exascale computer is expected to be put into operation in national supercomputing centers in Shanghai and Shenzhen, said its developer Dawning Information Industry Co. Ltd.

An exascale computer is able to execute a quintillion calculations per second. In China, prototypes are being developed by three teams led by the National Research Center of Parallel Computer Engineering and Technology (NRCPC), Dawning Information Industry, and the National University of Defense Technology (NUDT).

With Shuguang's launch, the three developers have all launched prototype exascale computing machines, marking a further step toward China's successful development of the next-generation supercomputer.

"The launch of a prototype exascale computing machine helps researchers test and improve key technologies through trial and error, and clear obstacles for the final computing system," said Zhang Yunquan, a researcher with the Institute of Computing Technology at the Chinese Academy of Sciences.

 

[JSCh]

China Reveals Third Exascale Prototype | TOP500 Supercomputer Sites
Michael Feldman | October 22, 2018 18:08 CEST

The Xinhua News Agency has reported that China has launched the prototype of Shuguang, an exascale supercomputer being developed by Dawning Information Industry, also known as Sugon.

According to the Xinhua report, the Shuguang machine is “expected to be put into operation in national supercomputing centers in Shanghai and Shenzhen.” The exascale prototype represents the third such systems put into operation over the last year. As we reported in August, the first two protypes were announced during this past summer.

The first of these, unveiled by National Supercomputing Center in Tianjin, is the forerunner to the Tianhe-3 exascale machine. It’s presumed to be an Arm-based supercomputer, based on Phytium’s Xiaomi platform. The second prototype is the precursor to the Sunway exascale system, which is set to be installed at the National Supercomputing Center in Jinan. The Sunway machine is expected to be based on a future version of the ShenWei processor, the latest version of which is used to power the 93-petaflop Sunway Taihulight supercomputer. However, no details were revealed about the nature of the processors powering the two prototypes.

The same is true for the third prototype announced this week, although the final Shuguang exascale machine is expected to rely on domestically-produced x86 processors. And now that Chinese chipmaker Hygon has such technology in the form of a Zen server CPU license, the most likely path to indigenous x86-based exascale machinery looks like it will be through AMD’s intellectual property. Hygon is already shipping their first locally-made Zen chips into the domestic market, under the name of “Dhyana,” and these could certainly be the basis for the Shuguang prototype.

Hygon’s EPYC license is restricted to the AMD’s first-generation Zen architecture, which is unlikely to be powerful enough to be the foundation of an exascale supercomputer. That suggests the Chinese are already looking ahead to future Zen designs to fuel their x86 needs. AMD’s Zen 2 EPYC chip, codenamed Rome, is already in the pipeline and is scheduled to go into production next year. But for exascale systems, the Zen 3 EPYC platform would be the most logical choice. That chip is on schedule for a 2020 release, which is just in time for China’s initial crop of exascale supercomputers.

Of course, Intel could also offer the Chinese licenses of its x86 technology and considering the escalating nature of the US-China trade war, it might behoove all chipmakers to strike local licensing deals to avoid running afoul of future retaliatory tariffs. On the other hand, the US government could decide to cut off Chinese licensing of advanced US technology based on national security concerns. That would effectively scuttle China’s plans to produce domestically produced x86-based supercomputers, not to mention more mainstream servers. But that’s getting pretty far ahead of ourselves.

The bottom line is that China has managed to launch all the prototypes for its three-pronged exascale strategy and do so at least two years prior to the deployment of the final systems. Whether China meets its 2020 target for at least one of these exascale machines remains to be seen, but having three options upon which to draw would seem to improve its chances.

 

[JSCh]

Three Chinese teams join race to build the world’s fastest supercomputer | Science | AAAS
By Dennis Normile
Oct. 24, 2018 , 1:45 PM

TIANJIN, CHINA—In a cavernous room just off the marble floored lobby of China's National Supercomputer Center of Tianjin stand more than 100 wardrobe-size black and gray metal cabinets, arranged in ranks like a marching army. They contain the Tianhe-1A supercomputer, which 8 years ago became the first Chinese machine to reign, briefly, as the world's fastest computer, running at 2.57 petaflops (or quadrillion floating point operations per second). But just upstairs from Tianhe-1A—and off-limits to visitors—is a small prototype machine that, if successfully scaled up, could push China to the top of the rankings again. The goal is a supercomputer capable of 1 exaflop—1000 petaflops, five times faster than the current champion, the Summit supercomputer at Oak Ridge National Laboratory in Tennessee.

China is vying with the United States, Europe, and Japan to plant its flag in this rarefied realm, which will boost climate and weather modeling, human genetics studies, drug development, artificial intelligence, and other scientific uses. But its strategy is unique. Three teams are competing to build China's machine; the Tianjin prototype has rivals at the National Supercomputing Center in Jinan and at Dawning Information Industry Co., a supercomputer manufacturer in Beijing. The Ministry of Science and Technology (MOST) will probably select two for expansion to exascale by the end of the year. The approach is a chance to spur innovation, says Bob Sorensen, a high-performance computing analyst at Hyperion Research in St. Paul. It "encourages vendors to experiment with a wide range of designs to distinguish themselves from their competitors," he says.

China may not be first to reach this computing milestone. Japan's Post-K exascale computer could be running in 2020. The United States is aiming to deploy its first exascale system at Argonne National Laboratory in Lemont, Illinois, in 2021. The European Union is ramping up its own program. China is aiming for 2020, but the date may slip.

Being first is not China's only goal, however. Having three competing teams will ensure broad-based technological advancement in computer chips, operating software, networking, and data storage technologies, says Meng Xiangfei, a physicist leading exascale application R&D for the center here. Building domestic capacity is particularly important for central processing units (CPUs) and specialized chips called accelerators, which boost a computer's performance. China relied on U.S.-made Intel CPUs for several generations of supercomputers, says Jack Dongarra, a computer scientist at the University of Tennessee in Knoxville, but in 2015, the U.S. government barred the export of certain chips for security reasons. That move "provoked the Chinese government to make a heavy investment" in processors, he says. All three exascale prototypes use chips made in China.

The three-team strategy also allowed MOST to share costs with regional governments, which hope that a leading edge supercomputer will spur technological development and lure institutes and businesses. Qian Depei, a computer scientist at Beihang University in Beijing who serves on MOST's exascale evaluation team, says the prototypes cost about $9 million each; MOST put up half and the rest came from local sources.

The prototypes have faced a battery of tests for speed, stability, and energy consumption plus trial runs of software from different application areas, but the results are "very secret," Meng says. The final budget is also unclear, though at the outset a governmental advisory committee estimated one exascale computer would cost 2 billion to 3 billion yuan ($288 million to $432 million).

Even after the two winners are announced, Qian says the third team will probably remain involved so the expertise they've acquired is not wasted. Scaling up the prototypes, which operate in the range of 3 petaflops, will mean interconnecting enough CPUs and accelerators to reach an exaflop, refining the liquid cooling systems needed to remove heat and improve efficiency, and perfecting the operating software needed for the massively parallel arrangement of processors to work together.

China once lagged in developing application software needed to do interesting science with supercomputers, but it has been catching up, Meng says. For the past 2 years, Chinese groups have won the Gordon Bell Prize presented annually by the Association for Computing Machinery for innovations in applying high-performance computing to science, engineering, and large-scale data analytics. Chinese scientists are now working on new applications, says Yang Meihong, director of the Jinan center. For example, going to exascale will allow a dramatic improvement in the spatial resolution of global atmospheric models, "which will be greatly significant for a deeper understanding of the mechanisms of climate change," she says.

The United States still dominates among the truly powerful supercomputers used for research, with 21 systems in the top 50 to China's two. But scientists play down the ranking's importance. "Having the top 500 No. 1 supercomputer—that's pretty good, but that's not the goal," Qian says. "The real measure should be what kind of new science we have as a result of these computers," Dongarra says.

 

[JSCh]

China expands supercomputer share in TOP500
Source: Xinhua| 2018-11-12 22:33:45|Editor: mmm


DALLAS, Nov. 12 (Xinhua) -- China expanded its share on a global list of the world's fastest supercomputers, according to a biannual ranking of the Top500 published Monday.

The number of supercomputers installed in China increased from 206 in June to 227 now, accounting for 45.4 percent of the total, according to the ranking.

The number of supercomputers that call the United States home, by contrast, continued to decline, reaching an all-time low.

John Dongarra, professor of Innovative Computing Lab with the University of Tennessee, told Xinhua that China is making large changes with 227 systems compared with 109 in the United States.

However, systems in the United States are, on average, more powerful, resulting in an aggregate system performance of 38 percent, compared to 31 percent for China.

The Top 10 supercomputers saw five U.S.-built systems with the first two captured by "Summit" and "Sierra," which had similar architectures with IBM cores and NVIDIA GPUs.

China's Sunway TaihuLight supercomputer is ranked third with its performance of 93.0 petaflops.

Tianhe-2A (Milky Way-2A), deployed at the National Supercomputer Center in Guangzhou, is now in the number four position, according to the ranking.

Germany had a new Top 10 entry with SuperMUC-NG, ranking number eight. It was built by Chinese tech company Lenovo, the top supercomputer manufacturer in the ranking, producing 140 of the total.

The most energy-efficient system on the Green500, a list released along with the Top500 to evaluate a system's energy efficiency, is once again claimed by the Shoubu system B at RIKEN scientific research institute in Japan.

The Top500 list is considered one of the most authoritative rankings of the world's supercomputers. It is compiled on the basis of machine performance on the Linpack benchmark by experts from the United States and Germany.

 

[TaiShang]

China with 45% of world's fastest supercomputers

Li Yue China Plus Published: 2018-11-13


China now holds 45% of the world's top 500 supercomputers, as well as occupying 2 places in top 10, according to the 52nd edition of the supercomputer TOP500 list released in Dallas, Texas, November 12, 2018.

The TOP500 list of supercomputers released by the TOP500 organization, November 12, 2018. [Photo: top500.org]

In the top 10, 5 of the super computers are from US Department of Energy (DOE), with the first two Summit at the Oak Ridge National Laboratory (ORNL) and Sierra at Lawrence Livermore National Laboratory. The Oak Ridge computer is the perennial champion.

China's Sunway TaihuLight supercomputer ranks 3rd on the list. TaihuLight was developed by China's National Research Center of Parallel Computer Engineering & Technology (NRCPC) and installed at the National Supercomputing Center in Wuxi, Jiangsu Province. Tianhe-2A (Milky Way-2A), deployed at the National Supercomputing Center in Guangzhou, Guangdong Province, is the number four on the TOP500 list.

Sunway TaihuLight supercomputer, deployed at the National Supercomputing Center in Wuxi, Jiangsu Province, May 23, 2018. [Photo: VCG]

On the whole, China's share of the TOP500 supercomputers continues to rise. There are now 227 Chinese supercomputers on the TOP500 list, up from 206 on the list six months ago.

Complied by the international organization "TOP 500", the list of the world's top supercomputers has been put out since 1993 and is released every six months.

http://chinaplus.cri.cn/news/china/9/20181113/209219.html

 

[JSCh]

China supercomputer manufacturer demonstrates its new energy-efficient system
Source: Xinhua| 2018-11-14 14:02:08|Editor: mym


DALLAS, Nov. 13 (Xinhua) -- Chinese supercomputer manufacturer Sugon demonstrated its new high performance computer called Silicon Cube at an on-going international supercomputer conference in Dallas.

Silicon Cube adopted the phase-change liquid cooling technology with the high energy efficiency, reducing the power usage effectiveness to 1.04.

Li Bin, general manager of High Performance Computing (HPC) Division at Sugon, told Xinhua at the conference named SC18 that only 4 watts power would be used to cool down equipment of 100 watts while 50 to 100 watts power might be used for an ordinary system.

As the supercomputers are eying a billion billion calculations per second benchmark, one challenge lies in its large power consumption and limits on system scale, according to Li.

"It's no longer possible to improve the performance of supercomputers by simply enlarging system scale or increasing power dissipation," said Li. "Higher performance has to depend more on lower power usage effectiveness."

Sugon scientists developed the immersion phase change liquid cooling technology. They soaked the mainboards with CPUs and GPUs into a specific fluids that may bubble up to air in a constant temperature, bringing away more heat than air-cooled technology while not causing short-outs of the circuits.

Also, the new computer applied the heterogeneous computing architecture, improving its supports for artificial intelligence applications, according to Sugon.

Ye Jian, Sugon's Chief Operating Officer, said that the popularity of artificial intelligence, largely represented by machine learning and deep learning, had boosted the explosive growth of computing resources demands, offering more potential for Silicon Cube's application in many industries.

Sugon is one of China's earliest and largest HPC vendors. It produced 57 supercomputers in the latest Top 500 supercomputer rankings.

 

[JSCh]

https://twitter.com/x/status/1063433244400447488

HPC Guru‏ @HPC_Guru
HPC Guru Retweeted Tsinghua University

Congratulations @Tsinghua_Uni for the clean sweep of Student Cluster Competitions this year! #SC18 #HPC

https://twitter.com/x/status/1063331692289638402

10:06 PM - 16 Nov 2018

3:23 PM - 16 Nov 2018