It is reported that with the 56th edition of the TOP500, Japan, the USA, Germany, Italy, and France have been just crossed the top 100 of the world’s most powerful supercomputers. Japan’s Fugaku has got the first position followed by IBM’s Summit and Sierra in the U.S. Fugaku has been equipped with new hardware to beef up the system, Fugaku grew the high-performance LINPACK benchmark to 442 petaflops, in comparison to the 416 petaflops the system that has got while it has been debuted in June 2020 with 6.4% speed improvement.
This HPL benchmark is indicating to the performance s of the dedicated system while solving a dense system of the equations that have been used to determine the Top500.
Here is the list of the top 10 supercomputers in the world:
10. Dammam-7, Saudi Arabia
It is the second new edition at the top of the list. It has been installed at Saudi Aramco in Saudi Arabia, the HPE Cray CS Storm systems have used Intel Gold Xeon CPUs and NVIDIA Tesla V100 GPUs. It has got 22.4 petaflops and is also the second commercial supercomputer in the top 10.
9. Frontera, US
It has been installed at the Advanced Computing Center at the University of Texas at Austin in September year, Frontera is a Dell C6420 system manufactured by Intel. It is using 448,448 of its Intel Platinum Xeon cores, the system has got 23.5 petaflops. Frontera has aided the research along with the areas of science containing those related to quantum mechanics, drug design, eradicating emerging viruses, also physics black holes.
It is a Dell PowerEdge system, HPC5 was installed by the Italian company Eni S.p, A and also has been located inside Green Data Center of Eni in Italy. It is one of the most powerful and sustainable systems in the world, HPC5 has been used to research the new origins of energy.
7. JUWELS Booster Module, Germany
It seems that Atos JUWELS Booster Module is the latest entrant on the list. It seems that the BullSequana machine had recently been installed at the Forschungszentrum Julich (FZJ) in Germany and also is one of the power systems in Europe, coming in at 44.1 HPL petaflops. It revolves around the modular system architecture, JUWELS has been powered by AMD processors and NVIDIA GPUs like the Selene system.
6. Tianhe-2A, China
It has 61.4 petaflops, Tianhe0-2A jumped one spot and get the sixth position. It has been developed by China’s National University of Defense Technology, it Installed at the National Supercomputer Center in Guangzhou.
Tinahe-2A has been charged by Intel Xeon CPUs and NUFT’s Matrix 2000 DSP accelerators. It will be utilized for simulation, analysis, and government security applications.
5. Selene, U.S.
It has been installed in-house vat NVIDIA Corp, Selene climbed up to the fifth position from the seventh position in June rankings. After the recent upgrade, Selene got 63.4 petaflops on HPL, and also close to double its previous score of 27.6 petaflops.
NVIDIA has revealed Selene, is an AI supercomputer this June this year, after the construction, and runs it in less than a month. It has the key to use contain system development and testing, in-house AI Workloads and also s chip design work.
4. Sunway TaihuLight, China
It has been installed as the national supercomputing center in Wuxi, Sunway TaihuLight, China has been held previously at the first position since two years last year it got the third position, now slip to the fourth position.
It has been created by China’s National Research Center of Parallel Computer Engineering & Technology, It has achieved 93 petaflops on the HPL benchmark. It has been powered by Sunway SW26010 processors.
3. Sierra, U.S.
It is a system at the Lawrence Livermore National Laboratory in California, it has got the HPL mark of 94.6 petaflops. It has 4,320 nodes manufactured with the two Power9 CPUs and also four NVIDIA Tesla V100 GPUs. It seems that Sierra also has been made it to the 15th position on the Green500 list as the world’s most energy-efficient supercomputer.
2. Summit, U.S.
It is mainly based on at the Oak Ridge National Laboratory in Tennessee, it was built by IBM and has the fastest system in the US, It was launched in 2018, it has a performance of 148.80 petaflops and also has 4,356 nodes, and each one housing with the two 22-core Power9 CPUs and six NVIDIA Tesla V100 GPUs.
Currently, the two teams are working on this and got the prestigious Gordon Bell Prize for the outstanding achievement with such high-performance computing.
1. Fugaku, Japan
It’s now the fastest supercomputer that has been built by Fujitsu, Fugaku is installed at the RIKEN Center for Computational Science in Kobe, Japan. There is such additional hardware that the system got a new world record of 442 petaflops as a result of HPL, and it is making three times before of the two system number list.
The director of RIKEN, Satoshi Matsuoka stated that the improvement came like they were finally being able to use the entire machine rather than just a good chunk of it.
Since the competition in June, the tea, has become able to fine-tune the code for the maximum performance, as Matsuoka says, ” I don’t think we can improve much anymore.”
A big thanks for the source