Top 5 Fastest Supercomputers In The World

Supercomputers are high-level computers used for managing a very large no of databases, great amounts of computation and it also has a high level of performance as compared to general-purpose computers.

The desire for high-performing fastest supercomputers in market research, scientific and engineering work in many fields, and other business-associated models continues to surge today. 

This is where a supercomputer comes in, performing at or near its current highest operational rate. The performance of a supercomputer is commonly measured in floating-point operations per second (FLOPS) instead of million instructions per second.

in this article, we discuss the top 5+ fastest supercomputers in the world by speed. Many types of supercomputers are available in the world.

Top 5 Fastest Supercomputers In The World

TOP 5+ SUPERCOMPUTERS IN THE WORLD
TOP 5+ SUPERCOMPUTERS IN THE WORLD

Here is the list of the fastest supercomputer in the world. the list is totally based on the speed and ability of the fastest supercomputers. While India's very own PARAM Siddhi-AI is the 62nd most powerful supercomputer in the world. let's discuss the top 5 fastest supercomputers in the world. 

 Top 5 Fastest Supercomputers In The World

1. Fugaku

The most recent fastest supercomputer and the best supercomputer in the world (until Finland gets its new HPE built) is Japan's Fugaku. The supercomputer is powered by Fujitsu's 48-core A64FX SoC, the first to reach number one using ARM. 

You can find it at RIKEN Center for Computational Science (R-CCS) in Kobe, Japan. It hits a lofty 415.5 petaflops. In 2020 Fugaku is the fasted supercomputer in the world with the 442 (PFLOPS) and also with the cost of 1213 million USD.

Name:- Fugaku
Model:- Supercomputer Fugaku
Manufacturer:- Fujitsu limited
Start Year:- 2020
Performances (PFLOPS):- 442(PFLOPS)
Memory:- HBM2 32 GiB/node
Storage:- 1.6 TB NVMe SSD/16 nodes (L1); 150 PB shared Lustre FS (L2); Cloud storage services (L3)
CPU/GPU Vendor:- Fujitsu
CPU:- A64FX
CPU Cores:- 158,976 × 48-core Fujitsu A64FX (ARM-based) @2.2 GHz
OS:- Custom Linux-based kernel
Cost:- US$1 billion (total programme cost)
Top 500 Ranking:- June 2020 to November 2021 1st

This list is based on the TOP500 Rmax measurement using the LINPACK benchmark at FP64 precision.

2. IBM's Summit.

Based at the Oak Ridge National Laboratory (ORNL) in Tennessee, Summit was built by IBM in 2018 and is the fastest system in the US. 

it has a performance of 148.8 petaflops and has 4,356 nodes, each one housing two 22-core Power9 CPUs and 6 NVIDIA Tesla V100 GPUs. in 2018 Summit is the 2nd fastest supercomputer in the world with the speed of 148 (PFLOPS) and the cost of 300 million USD.

Name:- Summit
Model:- IBM Power System AC922
Manufacturer:- IBM 
Start Year:- 2018 
Performances (PFLOPS):- 148(PFLOPS) 
Storage:-  250PB
CPU/GPU Vendor:- IBM, Nvidia
CPU:- POWER9, Tesla
CPU Cores:- 9,216 × 22-core IBM POWER9 @3.07 GHz
OS:- Linux (RHEL 7.4)
Cost:- 449 million USD (Approx.)
Top 500 Ranking:- 2nd

This list is also based on the TOP500 Rmax measurement using the LINPACK benchmark at FP64 precision.

3. Sierra

Sierra is one of the world’s super fastest supercomputers, which is developed for the Lawrence Livermore National Laboratory (LLNL) for use by the National Nuclear Security Administration (NNSA) as the 2nd Advanced Technology System. 

The system provides computational resources that are vital for nuclear weapon scientists to fulfill the NNSA’s stockpile stewardship mission through simulation instead of underground testing. 

The IBM-built Sierra supercomputer provides over 6 times the sustained throughput performance and more than 5 times the sustained scalable science performance of Sequoia, with a one hundred twenty-five petaFLOP/s peak. 

This supercomputer, which combines IBM’s Power 9 processors and NVIDIA’s Volta graphics processing units, is over five times more power-efficient than Sequoia, with a peak power consumption of approximately 11 MW. With 94 (PFLOPS) Sierra is the world's 3rd fastest supercomputer.

Name:- Sierra
Model:- IBM Power System S922LC
Manufacturer:- IBM
Start Year:- 2018
Performances (PFLOPS):- 94(PFLOPS)
Memory:- 2–2.4 PiB
CPU/GPU Vendor:- IBM, Nvidia
CPU:- POWER9, Tesla
CPU Cores:- 8,640 × 22-core IBM POWER9 @3.1 GHz
OS:- Linux (RHEL)
Cost:- 280 million USD 
Top 500 Ranking:- 3rd

IBM is developed sierra for the purpose of Nuclear weapon simulations.

TOP 5+ SUPERCOMPUTERS IN THE WORLD CODEPLACEWEB
TOP 5+  FASTEST SUPERCOMPUTERS IN THE WORLD

4. Sunway TaihuLight

It is not only the fastest it's also energy efficient. Sunway TaihuLight, a Chinese supercomputer, is ranked fourth in the TOP500 list, as of November 2018, with a LINPACK benchmark rating of 93 petaflops. 

The supercomputer uses a total of 40,960 Chinese-designed SW26010 manycore 64-bit RISC processors based on the Sunway architecture. Each processor chip carries 256 processing cores and an additional four auxiliary cores for system management (also RISC cores, just more fully featured) for a total of 10,649,600 CPU cores across the entire system. 

The computer, which runs the Linux-based Sunway Raise OS 2.0.5, uses the SW26010 processor, a chip designed by Shanghai High-Performance IC Design Center. in 2016 Sunway TaihuLight was the fasted supercomputer in the world with 93 (PFLOPS).

Name:- Sunway TaihuLight
Model:- Sunway MPP
Manufacturer:- NRCPC
Start Year:- 2016
Performances (PFLOPS):- 94(PFLOPS)
Memory:- 1.32 PB (5591 TB/s total bandwidth)
Storage:- 20PB
CPU/GPU Vendor:- NRCPC
CPU:- Sunway SW26010 
CPU Cores:- 40,960 × 260-core Sunway SW26010 @1.45 GHz
OS:- Sunway RaiseOS 2.0.5 (based on Linux) 
Cost:- 290 million USD
Top 500 Ranking:- 4th

Sunway TaihuLight was started in 2016 with the purpose of Oil prospecting, life sciences, weather forecast, industrial design, pharmaceutical research.

5. Perlmutter

Perlmutter (also known as NERSC-9) is a supercomputer scheduled to be delivered to the National Energy Research Scientific Computing Center of the United States Department of Energy in late 2020 as the successor to Cori. 

It is being built by Cray and is based on their upcoming Shasta architecture which is planned to utilize Zen 3 based AMD Epyc CPUs ("Milan") and next-generation Nvidia Tesla GPUs. 

Its intended use-cases are nuclear fusion simulations, climate projections, and material and biological research. It is predicted to reach approximately 100 PFLOPS of processing power. In 2021  Perlmutter is listed in the 5th position in the list of to500 fasted supercomputers.

Name:- Perlmutter
Model:- hp
Manufacturer:-  HPE
Start Year:- 2021
Performances (PFLOPS):-  63(PFLOPS)
Memory:- 256 GiB/node
Storage:- 35 PB, 5 TB/s Shared all-flash Lustre Filesystem
CPU/GPU Vendor:- HPE
CPU:- AMD Epyc 7763 
CPU Cores:- AMD Epyc 7763 64-core @2.45 GHz
OS:- Linux (HPE Cray OS)
Top 500 Ranking:- 5th

Perlmutter is the latest supercomputer .it starts in 2021 for the purpose of nuclear fusion simulations, climate projections, material, and biological research, and computational cosmology, and it is also listed in 5th position.

Please do not enter any spem link in the comment box.

Post a Comment (0)
Previous Post Next Post
close