Ukraine Snake Island Flag - Buy it Here!

Weapons of Mass Destruction (WMD)


Chinese Supercomputing

With a huge memory capacity and tremendous processing speed, supercomputers is capable of doing huge processing of data, including scientific, engineering analysis and military applications. They are used for a wide range of complex works such as weather forecasting, jetliners design, oil and gas exploration and nuclear explosions simulation. The processing speeds of a supercomputer are measured in floating point operations per second, or "FLOPS."

The history of supercomputing can be traced back to the United States with a series of computers at Control Data Corporation (CDC), designed by Seymour Cray to use innovative designs and parallelism to achieve superior computational peak performance. The CDC 6600, released in 1964, is generally considered the first supercomputer.

March 1956, Professor Min Nai, Professor Hu Shihua, Professor Xu Xiangyu , Zhang XiaoxiangProfessor Wu Jikang, associate professor Wu Peikang and Peking University party and government officials, participated in the International Conference on "Computing Technology Development Road" hosted in Moscow, and went to take lessons from the former Soviet Union to make technology for the computer part of China's 12-year plan ready. The then principal members of the delegation later participated in the 12-year plan. In addition, Fan Xinbi, Xia Peizhuo and Jiang Shifei also joined the planning and formulation. In the ensuing 12-year plan, it was confirmed that China wanted to develop a computer and approved the establishment of CAS Institute of Computing Technology, Semiconductor, Electronics and Automation and other four research institutes.

China's first large-scale semiconductor 103 computers (fixed-point 32 binary bits, 2500 operations per second) in the CAS Institute of Computing Technology was born and delivered in 1958. The backbone of research and development are Dong-Qiu ball, Wang Xingang and other young people. Subsequently, the first large-scale digital electronic computer 104 (floating point 40 bits, 10,000 operations per second), led by General Staff Professor Zhang Xiaoxiang, was also delivered in 1959.

Defense is the primary service object in any developed country, first of all for the development of computer defense services, used in the national strategic plan, and China was no exception. In 1958, Zhang Shilong from Peking University led the teachers and students of Peking University, including Wang Xuan, then a student, to work with the Chinese PLA Air Force to design and develop the digital computer "Beijing One" and deliver it to the Air Force for use. At that time, commander in chief Zhu De of the Chinese People's Liberation Army also personally visited the "Beijing No. 1".

Zhang Shilong led the teachers and students (including Wang Xuan and Xu Zhuoqun), and immediately put themselves into the "Red Flag" computer design work designed by Peking University. At that time, the goal set by the Soviet Union was higher than the 104 machine helped by the former Soviet Union experts in 1962 Trial trial success. However, due to the relocation and the Cultural Revolution, the "red flag" has not been able to resume and continue its work after the dislocation.

In 1964, Institute of Computing Technology , Chinese Academy of SciencesWu Ji Kang, Fan Xinbi designed the 119 machine (universal floating-point 44 binary, 50,000 times per second) is also delivered, this was China's first self-designed tube general-purpose computer, but also the world's fastest Tube computer . At that time, the United States and other developed countries had already moved into the area of ??transistor computers.

No. 109 B (32 floating-point bits at 60,000 operations per second), a self-designed transistor computer led by Jiang Shifei, Institute of Computing Technology, Chinese Academy of Sciences, was delivered in 1965. In order to develop the "two bombs and one satellite" project, in 1967, led by Jiang Shifei, a computer scientist at the Chinese Academy of Sciences, a computer designed specifically for two bombs and one satellite was designed and delivered. Two 109 C were installed in the Second Machinebuilding Ministry for nuclear bomb research and the Seventh Aircraft Institute for rocket research purposes. The 109C was used for 15 years and is known as the "Meritorious Computer". It was the first computer in China to have time-sharing, interrupt system and management procedures. The model 109C computer boasted a performance of 115 kiloflops, which paled in comparison with Control Data Corporation’s 6600 computer, the American machine that was then the world’s fastest computer — at three megaflops.

655 Computers were used for miniaturized nuclear weapons. In 1973 the East China Institute of Computing successfully developed the 655 Computer integrated circuit computer with 600,000 operations per second. The plant began work in 1969 on the 655 Computer, a redesigned TQ6-type computer. With language compiler, symbol assembler, interrupt and input and output management program, it achieveed program automation and multi-channel program running, with the first disk operating system. IT had a memory capacity of 131,072 words, three interrupts, and equipped with teletype machines, photoelectric machines, wide line printers, X-Y recorders, disk drives, tape drives, large plotters and other nearly 30 peripherals. In April 1974, the trial was successfully completed and it was put into production formally. By 1980, a total of 23 units were produced.

757 Vector Computer757 Vector Computer surpassed existing computers in meeting the needs of computational fluid dynamics, so in March 1973 Tsien Hsueshen asked the CAS Institute of Computing to develop new parallel supercomputers. In May 1973 they put forward the national production for the 757 vector computer (10M ECL circuit - a single pipeline) as a feasible scheme. But Chinese accounts relate that due to the serious interference in the Cultural Revolution and the unbridled leadership of the "lantern" after the Cultural Revolution, despite the successful development of integrated circuit chips for 20M low-power ECL circuits, "all device localization of one hundred million high-performance supercomputer" research and development, because the mission changes, eventually ran aground.

Vector computing is very suitable for the structural characteristics of the pipeline computer. The combination of vector-based parallel computing and pipeline architecture can largely overcome the shortcomings of too much instruction processing capacity, uneven storage access, serious waiting and poor running flow in the conventional pipelined computers and give full play to the parallel processing structure Potential, significantly improving computing speed.

The 757 Computers could solve the large-scale science, engineering calculation problem. Chinese Academy of Sciences and more than 30 departments and regions more than 80 units worked together, and in November 1983 approved the appraisal, and delivered the computer to the Ninth Institute for use. The components and equipment were based on domestic research, with independently developed vector processing and multi-vector accumulator concepts. The operation speed was 10 million times per second, scalar operation speed of 2.8 million times per second. Complete peripherals included more complete system software, FORTRAN77 compiler and vector function expansion. Multi-channel operating system features used calibration, correction, re-calculation and diagnosis techniques.

The State Evaluation meeting of the "757" 10-MIPS [million instructions per second] computer system was conducted from 13-15 November 1983 in Beijing. The meeting was carried out by a State Appraisal Board authorized by the State Council's Computer and Large-Scale Integrated Circuit Leading Group. The session was attended by 120 representatives of 50 units nation­ wide. Comrade Fang Yi [2455 3015], State Council member and State Scienti­ fic and Technical Commission chairman, and CAS Director Comrade Lu Jiaxi [4151 0857 6932], spoke at the meeting.

The State Appraisal Board, consisting of 26 eminent Chinese computer experts and leadership personnel from the various departments concerned, worked conscientiously at the conference. It heard the researchreport delivered by Comrade Wu Jikang [0702 0415 1660] on behalf of the CAS Institute of Computing Technology, which developed the "757," the technical evaluation report made by Comrade Jin Yilian [6855 1837 3425] on behalf of the Technical Evaluation Group, and reports by four subgroups of the Technical Evaluation Group. The board also studied all related documentation and tech­ nical data. Ultimately, it unanimously granted state certification for the 10-MIPS 757 computer system and held a signing ceremony.

The Model 757 Computer, the first large-scale vector digital computing system self-developed by China, was honored with the CAS Special Prize for Science and Technology Progress and the National First Prize for Science and Technology Progress in 1984 and 1985 respectively.

Galaxy-I Mega-ComputerGalaxy-I Mega-Computer met the Second Generation Strategic Missile and Nuclear Weapon Development requirements. China's first giant computer, named "Galaxy" was born in the National University of Defense Technology in November 1983, after 5 years. Its successful development declared to the world that China had become a country capable of independently designing and manufacturing supercomputers after the United States, Japan and other countries. The 77th Annual Report of the State Commission of Science and Technology approved the 100 megaFLOPS project in November 1984. Four machine builders, Chinese Academy of Sciences, Shanghai Municipality, Jiangsu Province and Heilongjiang Province were proposed at the time. Fang Yi presented the research organization, in March 1985 to Deng Xiaoping, who listened to report. Changsha Institute of Computer Research Institute was responsible for the development. Deng Xiaoping said to Zhang Aiping: "Now that the supercomputer is complete, you have to establish a military order."

The elements, devices and external equipment were all domestically produced. Shen Ya Cheng was responsible for 20M low-power ECL circuit integrated circuit chip. Galaxy was said for the first time in the world used a dual CPU program, which Chinse sources claim was stolen by large US companies. A total of 3 sets was produced: petroleum, western computing center and university computer institute. The development of supercomputer performance followed the Thousand Times Law, which means that performance will increase by 1000 times every 10 years. In fact, the development speed may be faster.

Galaxy-II Military Mega ComputerGalaxy-II Military Mega Computer was developed by Jiangnan Institute of Computing Technology starting in December 1985. In November 1991, the first scalar billion giant computer system was successfully completed. In the 1980s the mega-computer mission was assigned to the National Defense Science and Technology University, Academy of Sciences to study the application of the super-smart machine. At the same time it carried out super small and medium-sized machine series of design and development work and humanoid machine translation research. In 1986 the giant supercomputer was included in the 863 plan. Due to the very slow progress and the rapid development of VLSI in the world, it was foreseen that in a few years, the originally planned cost of US$20,000 will create a computer with an international price of US$200,000 and become a provider of performance and Computer equivalent of a $200 chip. The development of the Galaxy-II had been relatively slow as compared with progress in the Western high-performance computing industry during the decade. The Galaxy-II is, according to published specifications, a four-processor computer with a clock rate of 20 nanoseconds (equivalent to 50 megahertz] - slower even than a 1978-vintage Cray-I (12.5 ns), and far behind current high-end Japanese and Cray systems (which are in the 4-ns area). According to the Chinese press, the Galaxy-II had a total shared main memory size of 256 megabytes - better than a Cray-I but much less than current Western supercomputers. The Chinese state that the Galaxy-IJ has two independent 10-megabit-per-second input/output (1/0) subsystems - which if true is very slow and likely to be a significant limitation for users who need to move large data sets through the system. Overall, the Chinese have claimed that the Galaxy-IJ can perform 400 million 64-bit floating-point operations per second. The Galaxy-II's performance thus fell far short of current­ generation Western supercomputers; it was comparable to that of today's high-end scientific/engineering workstations, now available in the West/or under $100,000.

By 1994 nuclear weapons related work took place on a Galaxy-II computer at the Beijing National Meteorological Center [NMC]. The scientists from IAPCM could have been running test programs and/or portions of their nuclear modeling hydrocodes on the NMC Galaxy-II to gain experiernce with the computer before their institute received its own machine. This was the most likely scenario - it was precisely what US weapon designers often did to learn in advance how to optimize use of a new computer system before they take delivery and can run full nuclear simulations in their own secure facilities.

The Tianhe-1 supercomputer can complete an average of 1,400 online tasks on a daily basis, such as supercomputing for visual analysis and processing. The Tianhe-1, unveiled in 2010, is capable of at least 1 quadrillion calculations per second. It has served in a number of fields including digital mining research and development, animation, online finance, and big data, as well as e-governance of the Binhai New Area in Tianjin.

On 17 June 2013 the world supercomputer organization TOP500 officially announced the 41th TOP500 list of world supercomputers at the 2013 International Supercomputing Conference in Leipzig, Germany. Tianhe-2 supercomputer system developed by National University of Defense Technology heads the list again after Tianhe-1 became the first Chinese supercomputer champion in 2010. The Tianhe-2,a supercomputer developed independently by the National University of Defense Technology (NUDT) of the Chinese People’s Liberation Army (PLA), took the first place in the list with its peak computing speed of 54.9 petaflops per second and sustained computing speed of 33.9 petaflops per second.

When Tianhe-1A, Tianhe-2’s predecessor, ranked the first in the list of the world’s fastest supercomputers in November 2010, they made the judgment of “three remains” in the world’s supercomputing development, namely, the Western World’s superior position in the information technology field remained, the dominant position of the United States in the supercomputing field remained, and the situation of the world powers’ competing for the leading position in the development of supercomputers remained.

The 2017 edition of the semiannual T0P500 list of supercomputers showed that China's Sunway TaihuLight and Tianhe-2 are still the world's fastest and second fastest machines, but America's Titan was squeezed into the fourth place by an upgraded Swiss system. A supercomputer is a computer that has a much higher processing capacity, particularly a much higher speed of calculation, compared to a general-purpose computer.

A supercomputer-based simulation platform for the research and development of vehicles was launched in December 2018 at the National Supercomputer Center in north China's Tianjin. Supported by China's first petaflop supercomputer the Tianhe-1, the platform can help researchers and engineers develop key components of vehicles.

A car is composed of thousands of parts and fittings. A computer simulation can complete around 80 percent of the work in design and research of automobiles. International automakers like BMW, Volkswagen, and General Motors have founded their own supercomputer-based simulation platforms. The center has established cooperative relations with automakers and institutions such as China Automotive Technology and Research Center Co. Ltd., Foton Motor Group and Tianjin FAW. Tianjin FAW was the first to try the platform, which has optimized the company's research and development process for two types of its self-designed cars. Using the supercomputer, the company shortened its research time by 80 percent, saving more than 7 million yuan (around 1 million US dollars) in costs.

The National Supercomputer Center has completed the development of China's new-generation exascale supercomputer the Tianhe-3. The final version of the Tianhe-3 will be operational by 2020 when it is expected to be 200 times faster and have 100 times more storage capacity than the Tianhe-1.




NEWSLETTER
Join the GlobalSecurity.org mailing list



 
Page last modified: 19-12-2018 14:27:30 ZULU