Top 10 World’s Most Powerful Super Computers in the World

 Top 10 World’s  Most Powerful Super Computers

Top 10 World’s  Most Powerful Super Computers

 
Japan’s Fugaku mainframe computer has preserved its position because the much loved computing system, followed by IBM’s Summit and Sierra within the U.S.
In the 56th edition of the TOP500, Japan, USA, Germany, Italy and France have dominated the highest one hundred of the world’s most powerful supercomputers. Japan’s Fugaku mainframe computer preserved its position because the much loved computing system, followed by IBM’s Summit and Sierra within the U.S.
The Top500 project tracks the foremost powerful supercomputers within the world and is revealed biannually. The non-profit organization discharged its Gregorian calendar month 2020 results last week. This year, whereas 2 new systems created it to the highest ten, the total list recorded the littlest variety of recent entries since the project’s beginning in 1993.
As a result, Top500 representatives instructed that this list reflects a flattening performance growth curve.
What the results show
The world’s supercomputers area unit no longer solely quicker, however conjointly smarter, and support a larger form of workloads.
Nearly seventieth (140) of the systems on the TOP500 list, together with eight of the highest ten, area unit steam-powered by NVIDIA technology and area unit “increasingly mistreatment AI to assist researchers create discoveries quicker.”
NVIDIA’s end-to-end HGX AI supercomputing platform helps accelerate scientific computing, information analytics, and AI workloads.
“The shift to incorporating AI into HPC (High-Performance Computing), and a platform that extends on the far side ancient supercomputing centers, represents a big amendment in a very field that, since Seymour Cray’s bureau 6600 was launched in 1964, has targeted on harnessing ever larger, additional powerful machines for compute-intensive simulation and modeling,” NVIDIA aforementioned in a very journal post.
In October this year, Italian inter-university syndicate CINECA had proclaimed its plans to create the world’s quickest AI mainframe computer. The new “Leonardo” system is predicted to deliver ten exaflops of FP16 performance in AI workloads. it'll feature around fourteen,000 Nvidia A100 GPUs and around three,500 Intel Sapphire Rapids CPUs.
In the case of a mainframe computer, its performance capabilities area unit indicated employing a commonplace rate referred to as FLOPS or floating-point operations per second. This rate indicates the amount of floating-point arithmetic calculations a system will perform per second. whereas a one petaFLOPS (PFLOPS) computing system will perform one quadrillion (10^15) floating-point operations per second, one exaFLOPS (EFLOPS) computing system will perform one large integer (10^18) floating-point operations per second.
Meanwhile, Intel has remained the well-liked processor supplier, with ninetieth of the systems equipped with its Xeon or Xeon letter of the alphabet chips.

With 212 systems on the list, China has exerted unequivocal domination over the list, followed by the U.S. with 113 systems, and Japan with thirty four. However, once it involves combination performance, the U.S. continues to guide at 668.7 petaflops, compared to China’s 564.0 petaflops. Meanwhile, Japan has associate degree combination performance of 593.7 petaflops.

Supercomputers, that have the unfortunate aspect result of intense monumental amounts of energy, area unit currently conjointly hierarchal on the Green500 by energy potency. On the Green500 list, the new NVIDIA DGX SuperPOD within the U.S. was found to be the foremost energy-efficient system. it had been followed by MN-3 in Japan, and Atos-built JUWELS Booster Module in Federal Republic of Germany.

 

 

Top 10 supercomputers

Notable developments within the high ten embrace the addition of 2 new systems, and a brand new performance benchmark set by the Fugaku mainframe computer.

Due to the addition of latest hardware to make stronger the system, Fugaku grew its High Performance LINPACK (HPL) benchmark to 442 petaflops, compared to the 416 petaflops the system achieved once it debuted in Gregorian calendar month 2020 – a six.4% speed improvement.

The HPL benchmark, that indicates the performance of a fanatical system once finding a dense system of equations, is employed to work out the Top500.

1. Fugaku, Japan

Built by Fujitsu, Fugaku is put in at the RIKEN Center for procedure Science (R-CCS) in Kobe, Japan. With its extra hardware, the system achieved a brand new record of 442 petaflops result on HPL, creating it thrice before the quantity 2 system within the list.

RIKEN’s director, Satoshi Matsuoka, explicit  that the advance came as they were “finally having the ability to use the whole machine instead of simply a decent chunk of it.”

Since the Gregorian calendar month competition, his team has been ready to fine-tune the code for optimum performance. “I don’t suppose we are able to improve a lot of any longer,” Matsuoka aforesaid.

2. Summit, U.S.

Based at the Oak Ridge National Laboratory (ORNL) in Tennessee, Summit was engineered by IBM and is that the quickest system within the United States.. Launched in 2018, it's a performance of 148.8 petaflops and has four,356 nodes, each housing 2 22-core Power9 CPUs and 6 NVIDIA Tesla V100 GPUs.

Recently, 2 groups engaged on Summit won the distinguished Gordon Bell Prize for outstanding accomplishment in superior computing, normally mentioned because the ‘Nobel Prize of supercomputing.’

3. Sierra, U.S.

A system at the Lawrence suffragist National Laboratory (LLNL) in Calif., Sierra has AN HPL mark of ninety four.6 petaflops. With every of its four,320 nodes equipped with 2 Power9 CPUs and 4 NVIDIA Tesla V100 GPUs, it's AN design like that of Summit.

Sierra additionally created it to the fifteenth position on the Green500 List of the world’s most energy-efficient supercomputers.

4. Sunway TaihuLight, China

Installed at China’s National Supercomputing Center in Wuxi, Sunway TaihuLight antecedently command the quantity one spot for 2 years (2016-2017). However, its rank has since fallen. whereas it had been in third position last year, it's currently slipped to fourth.

Built by China’s National center of Parallel pc Engineering & Technology (NRCPC), it achieved ninety three petaflops on its HPL benchmark. it's supercharged completely by Sunway SW26010 processors.

5. Selene, U.S.

Installed in-house at NVIDIA corporation, Selene jumped to fifth position from seventh position within the Gregorian calendar month rankings. once a recent upgrade, Selene achieved sixty three.4 petaflops on HPL, nearly doubling its previous score of twenty seven.6 petaflops.

NVIDIA unveiled  Selene, its AI mainframe computer, in Gregorian calendar month this year, once constructing and running it in but a month. Its key uses embrace system development and testing, in-house AI workloads, and chip style work.

6. Tianhe-2A, China

With 61.4 petaflops, Tianhe-2A (aka MilkyWay-2A) climbed one spot to sixth position. Developed by China’s National University of Defense Technology (NUDT), it's put in at the National mainframe computer Center in Kwangchow.

Tianhe-2A is supercharged by Intel Xeon CPUs and NUDT’s Matrix-2000 DSP accelerators. it'll be used for simulation, analysis, and certificate applications. It command the quantity one spot from Gregorian calendar month 2013 to November 2015.

7. JUWELS Booster Module, Germany

Atos-built JUWELS Booster Module is that the newest entrant on the list. The BullSequana machine was recently put in at the Forschungszentrum Jülich (FZJ) in FRG and is that the most powerful system in Europe, returning in at forty four.1 HPL petaflops.

Based on a standard system design, JUWELS is supercharged by AMD processors and NVIDIA GPUs, like the Selene system.

8. HPC5, Italy

A holler PowerEdge system, HPC5 was put in by the Italian company Eni S.p.A and is found within Eni’s inexperienced knowledge Center in European country. one amongst the foremost powerful and property computing systems within the world, HPC5 is employed to analysis new sources of energy.

With a performance of thirty five.5 petaflops, it's the foremost powerful system within the list used for business functions at a client website. It uses Intel Xeon Gold CPUs and NVIDIA Tesla V100 GPUs.

9. Frontera, US

Installed at the Advanced Computing Center at the University of Texas at capital of Texas in September year, Frontera could be a holler C6420 system equipped by Intel. Using 448,448 of its Intel noble metal Xeon cores, the system achieves twenty three.5 petaflops.

Frontera aids analysis across all areas of science together with those associated with quantum physics, drug style, eradicating rising viruses, and therefore the physics of black holes.

10. Dammam-7, Asian country

Dammam-7 is that the second new addition at the highest of the list. put in at Saudi Aramco in Asian country, the HPE Cray CS-Storm systems uses Intel Gold Xeon CPUs and NVIDIA Tesla V100 GPUs. It achieves twenty two.4 petaflops and is that the second business mainframe computer within the prime ten.