The most Powerful Supercomputers in the world
Since its inception, computers have shown a surprising evolution, reaching everywhere and deployed and used in almost all companies, automobiles and of course our homes.
Today, we have the so-called supercomputers that are machines with surprising speeds and very different infrastructures that we often use.
Their tasks are dedicated, aerospace research, monitoring of storms and natural conditions among other uses. This is the top 10 of the most powerful and fastest supercomputers in the world.
Position Number 10
Shaheen II, King Abdullah University of Science and Technology, 5,537,000 gflop/S.
Fujitsu has inhabited the TOP500 since its inception in 1993, and 1995 was not the exception, with three of the top 10 places. Japan’s Atomic Energy Research Institute has fallen on the list since 2008, even though it may be back soon, with the announcement of an agreement to buy a Silicon Graphics ice X system with a theoretical maximum speed of 2.4 petaflops, which would place it near the top 25 on the most recent list.
9. Hazel Hen
This supercomputer is installed in the HLRS – Höchstleistungsrechenzentrum (High Performance Computing Center) Stuttgart, in Germany and the computer is built on Cray XC40, Xeon E5-2680v3 12C 2.5GHz, and Aries interconnect (similar to the Shaheen II) and was also developed by Cray Inc.
Hazel Hen has 185,088 processing cores, even though it’s less than Shaheen processor cores, it is said to have performed better than the Shaheen II at Rmax of 5,640.2 TFlop/s, Rpeak of 7,403.5 TFlop/s. Its power usage was not also disclosed.
8. Piz Daint, National Supercomputing Center of Switzerland, 6,217,000 Gflop/S.
The Piz Daint supercomputer is located in the Swiss National Supercomputing Centre and was launched in April 2013. Piz Daint has gone from having 12 computer booths to 28, which helps its theoretical peak yield to the 6.3 petaflops.
This supercomputer is developed by the Los Alamos National Laboratory and Sandia National Laboratories for the U.S. Department of Energy’s National Nuclear Security Administration (NNSA) and it’s installed in Los Alamos, New Mexico.
The developer consists of Cray XC40, Xeon E5-2698v3 16C 2.3GHz, Aries interconnect. Trinity has 301,056 processing cores and Rmax of 8,100.9 TFlop/s, Rpeak of 11,078.9 TFlop/s and its power usage is undisclosed.
6. Mira, the National Laboratory of Argonne, 8,586,600 Gflop/s
It was created by the Argonne National Laboratory, in the U.S. and has 78,000 cores processors, being able to do up to 10 thousand quadrillion calculations per second.
The purpose of its development is to understand the movement of the galaxies and to follow the movement of the particles that form our universe.
5. Fujitsu K Computer, Advanced Institute of Computational Science RIKEN, 10,510,000 gflop/S.
Created by the Fujitsu Computer Group and a Japanese research institute. Total 88,128 processors working in concert, achieved a mark of 10.51 petaFLOPS (10,510 trillion floating point operations per second). It is used for climate research, meteorology, disaster prevention and soon in medicine.
4. Sequoia, Lawrence Livermore National Laboratory, 17,173,200 Gflop/S
It is located in the state of California, United States, I bm Sequoia belongs to the type of IBM Blue Gene/Q computers which has a performance of 16.3 petaflops per seconds (floating-point operations per second, unit of measurement of the performance of a computer).
From this speed, the IBM Sequoia supercomputer is able to calculate in an hour the same as about 6,700,000,000 people with a calculator over a period of time of 320 years.
3. Titan, Oak Ridge National Laboratory, 17,590,000 Gflop/s
Developed by Cray Inc., an American company dedicated to the manufacture of this type of technology. Cray Titan has 299,008 cores, with 18,688 AMD Opteron CPUs and has the same amount of GPUs Tesla K20x for each microprocessor.
The resulting size of RAM memory is more than 710 TIB, with 10 PB of storage and has more than 10,000 hard disks of 1 TB at 7200 rpm. Designed to study natural phenomena by developing virtual models, for example, of storms (hurricanes), molecules physics or any scientific requirement (such as those required by medicine).
2. Tianhe-2, China National Supercomputing Center, 33,862,700 gflop/S.
Developed by the National University of Defense Technology of China (NUDT) and the Chinese company Inspur. It has a performance of 33.86 petaFLOPS (33,860,000,000,000,000 floating-point operations per second), with a theoretical peak of 54.9 petaflops that makes it the fastest supercomputer in the world.
It is equipped with 16,000 nodes, each with two Intel Xeon processors Ivybridge E5-2692 (12 cores, 2.2 GHz) and three Intel Xeon processors Phi 31S1P (57 cores, 1.1 GHz), whose combination gives a total of 3,120,000 computing cores.
It is capable of storing 12.4 PB, has a system memory of 1,375 TIB (1.34 GDP) and uses the operating system Kylin Linux. It occupies an area of 720 m² and is estimated to have cost between 200 and 300 million of dollars.
No 1. The 93 petaflop Sunway TaihuLight
It is no more news that China has overtaken US in the top 500 best efficient supercomputers in the world. This supercomputer is established at the National Supercomputing Centre in Wuxi.
The computer can perform 93,000 trillion calculations in a second, which is its peak. This is faster and efficient in operation more than Tianhe-2 and can process 100% of information more than Tianhe-2. It’s mostly used in huge manufacturing industries for huge data analyses, forecasting weather etc..
Its processing cores and more than 10 million and almost 50,000 nodes which runs on Linux based OS. Presently, the 2 best supercomputer are now installed in China, built by the Chinese computer engineers while US still has 3 supercomputers in the 5 best supercomputers in the world.