Going back in time, back in the 60 years the first supercomputer was being launched, the Atlas, which was installed at the University of Manchester, and is much less powerful compared to a desktop computer that almost everyone has at home these days. Since the 60s, we can see that technology is making great leaps. These changes used to take a long time to feel, but because we are at a very high pace of technological development, we see that every 5 years the supercomputers become obsolete, thanks to this great advance in information and development of computers and technological research.
Today's Supercomputers are measured in PetaFLOPS, a processing speed equal to one million billion, or one billion trillion, floating point operations per second. These supermachines serve to aid research by scientists, meteorologists in weather forecasting or global warming, to stimulate brain activities or effects of global warming, technological advances and nuclear safety, and many other things worldwide.
According to a senior scientist at the Beijing Computing Center, he says that a small fraction of these supercomputers are currently being used. He claims that a supercomputer, no matter how fast it is today, will become garbage in five years.
Despite this assertion, nations are still investing billions of dollars a year in the race to advance technology and become the best players in the world of supercomputer technology. The speed of these supercomputers varies greatly, so we will rank them at exorbitant prices.
10. IBM Roadrunner (United States) - $ 130 million
The Roadrunner was built by IBM for the Los Alamos National Laboratory in New Mexico, USA. It became only operational in 2008, it was designed to have maximum petaFLOPS 1,7 performance. On 25's May day of 2008, it reached 1,026 petaFlops, making it the first system in the world to reach 1.0 petaflops in the TOP500 Linkpack. And in November of the same year he achieved a superior performance of 1,456 petaFLOPS, keeping his place at the top of the TOP500 list.
According to the Supermicro Green500 list at 2008, Roadrunner was chosen as the fourth largest energy-efficient supercomputer in the world. In 31 of March of 2013 the supercomputer was disassembled and replaced by the supercomputer called Cielo, this one in turn is more energy efficient.
9. Vulcan BlueGene / Q (United States) - $ 100 millions
Vulcan is an 24-rack supercomputer that was created by IBM for DoE and is parked at Lawrence Livermore National Laboratory in Livermore, California, USA. It is currently the fifth fastest supercomputer in the world with a peak of 5 petaflops, according to Top500.org. BlueGene / Q is the third generation of IBM projects (after BlueGene / L and BlueGene / P), which aims to develop supercomputers that can achieve petaFLOPS speed and low power consumption.
Vulcan has become functional at 2013 at the Lawrence Livermore National Laboratory for research in biology, plasma physics, climate science, molecular systems, solid and fluid engineering, and other complex subjects of study. It is also used to support the missions of DoE and the National Nuclear Security Administration (NNSA).
8. SuperMUC (Germany) - $ 111 millions
SuperMUC is currently the world's 14th fastest supercomputer, 2013 it was the tenth fastest supercomputer in the world, but with the speed of technological evolution, it was soon over. Plus he ranks second in supercomputers in Germany. SuperMUC works is operating at the facilities of the Leibniz Supercomputing Center (LRZ) at the Bavarian Academy of Sciences, he found installed near Munich.
Developed by IBM, it runs the Linux system, contains more than 19,000 Intel and Westmere-EX processors, and has a peak performance of more than 3 petaFLOPS. The system known for its form of cooling, a new discovery by IBM, called Aquasar, which uses hot water to cool processors. The project reduces the use of electricity in cooling by 40%.
SuperMUC is used by European researchers in a wide range of fields, including medicine, astrophysics, quantum chromodynamics, computational fluid dynamics, biological sciences, computational chemistry, genome analysis, and earthquake simulations.
7. Trinity (United States) - $ 174 millions
This Trinity supercomputer was commissioned by the United States government to keep its nuclear arsenal safe. The United States government has offered supercomputer manufacturer Cray a contract of 174 million dollars to build this Cray XC supercomputer, along with a Cray Sonexion storage system for the National Nuclear Security Administration (NNSA), in order to maintain a safe and secure security of its nuclear arsenal.
6. Sequoia BlueGene / Q (United States) - $ 250 million
Once again IBM enters the supercomputers game and once again the NNSA behind this order. IBM developed the BlueGene / Q Sequoia supercomputer for the NNSA as part of an Advanced Simulation and Computing Program. It was deployed in June of 2012 at the Lawrence Livermore National Laboratory, and it immediately became the fastest computer in the world, according to TOP500.org. It currently ranks third on the installation site, with a theoretical peak of 20 petaFLOPS, or 20 trillions of calculations per second.
Sequoia was the first supercomputer to exceed 10 petaFLOPS in sustaining performance, and some scientific record-measuring applications were run on the system. For example, the Cardioid code - a project model that performs the electrophysiology of the human heart - achieved almost 12 petaFLOPS with a real-time simulation. The computer is also used for studies of astronomy, energy, the human genome, climate change, and of course, nuclear weapons.
5. ASC Purple and BlueGene / L (United States) - $ 290 millions
These two supercomputers came as a single package. The two computers were announced by the DoE in 2002, in a contract with IBM for 290 million dollars. They were installed in 2005 at the Lawrence Livermore Laboratory, and were decommissioned in 2010. At the time, ASC Purple ranked 66th on the TOP500 list of supercomputers. And the BlueGene / L was of an older generation, and a lesser model than the BlueGene / Q, the system currently has four different supercomputers in the TOP 10 list of TOP500.
ASC Purple at Lawrence Livermore Laboratory was built as the fifth phase of the US Department of Energy and NNSA in the Advanced Computing and Simulation Program, built to simulate and replace the WMD - Weapons of Mass Destruction live test experiment. of Mass Destruction). BlueGene / L was focused on important scientific areas, such as the prediction of global climate change and studying the interaction between atmospheric density and pollution.
At a press conference, the DoE announced that these two systems would have 1,5 times more processing power than all the other 500 machines jointed from the 500 TOP2002 list.
4. Sierra and Summit (United States) - $ 325 millions
This time IBM is not alone, in collaboration with Nvidia, they join forces to help America regain its position in the top of the world's fastest supercomputers, technology breakthroughs, scientific research, and economic and national security. Built using IBM Power Servers and Nvidia Tesla GPU accelerators, the two Sierra and Summit supercomputers will be installed on 2017.
At present, the Tianhe-2 China supercomputer is the world's fastest computer, able to achieve 55 petaFLOPS power, which is twice as much as the second on the list. The next Sierra system will have no problem running more than 100 petaFLOPS, while the Summit will have up to 300 petaFLOPS processing capability.
The use of the Sierra at Lawrence Livermore National Laboratory, there comes the same story, will be to ensure the safety and efficiency of the country's nuclear program. While the Summit for scientific applications around the world.
3. Tianhe-2 (China) - $ 390 million
As mentioned above, Tianhe-2 ("Milky Way-2" in English or "Milky Way-2") is the current fastest supercomputer in the world. The Tianhe-2 was developed by a team of 1.300 scientists and engineers, and it is located in the National Supercomputer Center in Guangzhou. And since 2013 it has had consecutive rankings like #1 on the TOP500 list of faster supercomputers. Tianhe-2 was sponsored by the 863 High Technology Program which was initiated by the Chinese government in partnership with the government of Guangzhou province.
It was built in China by the National University of Defense Technology (NUDT) after the United States government rejected Intel's application for certification for exports of CPUs and coprocessor boards. This was a strategic move by the United States government, and a blow to Intel and its suppliers. This attitude prompted the Chinese industry to develop its own processor.
The Tianhe-2 is capable of performing 33.860 billion calculations per second. An hour of calculations on this supercomputer is the equivalent of 1,000 years of difficult sums by 1,3 billion people. Exceptional this number, and we see the brighter and brighter future for information technology. The Tianhe-2 is used for simulation applications, analyzes, and government security applications.
2. Earth Simulator (Japan) - $ 500 millions
Earth Simulator (strange name?) Was developed by the Japanese government in 1997. Costing the state coffers around 60 billion yen (yen), or about 500 million dollars these days. This supercomputer was developed in parallel as a vector system, running global climate models, and to evaluate the effects of global warming and the problems in solid earth geophysics.
Earth Simulation (ES) was completed at 2002 and was developed for the Japan Aerospace Exploration Agency, the Japan Atomic Energy Research Institute, and the Japan Marine Science and Technology Center .
Earth Simulation was the fastest computer in the world between 2002 and 2004. With the rapid growth of technology since then, it no longer holds a candle close to modern supercomputers, but it was great news early in the 21 century. The ES has several features to help protect the computer against earthquakes (rubber supports in a seismic insulation system) and spokes (a high voltage, shielded nest that hover over the building)
1. Fujitsu K (Japan) - $ 1.2 billion
Japan despite having the two most expensive supercomputers in the world. The K computer, named after the Japanese word "sei", is the fourth fastest supercomputer in the world, with a speed of 11 petaflops. The system costs 140 billions of yen (yen) or 1,2 billions of dollars to be created.
In the year of 2011, TOP500 ranked the K as the fastest supercomputer in the world, and in November of 2011, the system became the first computer to top 10 petaflops officially. And in the year of 2012, K was replaced by IBM's Sequoia as the world's fastest supercomputer. OK is located at RIKEN, Advanced Institute of Computational Science, it is 60 times faster than Earth Simulation (ES). OK costs 10 millions of dollars a year to be in operation, using 9,89 MW of power, or the equivalent of almost 10.000 suburban homes, or one million desktop linked computers.
