Top Ten Fastest Supercomputers 2018 Updated list

Top Ten Fastest Supercomputers 2018 Updated list

On our regular Bits And Screw's Top Ten List, Today we will be ranking the Top Ten fastest Supercomputers in the World for 2017 ranging from the Sunway TaihuLight to the Vulcan.

Last updated on June 26th, 2018 at 05:50 am

What is a Supercomputer


World’s Top Ten Fastest Supercomputers – June 2018 Updated list

 


# System Entity Processor TFlop/s (Peak) Power (kW)
1 Summit IBM & US Department of Energy

United States

IBM Power9 processors + 6 x Nvidia Tesla V100 GPU

2,282,544 Cores

2,801,664 GB RAM

187,659 TFlop/s 8,805.50 kW
2 Sunway TaihuLight

 

National Supercomputing Center in Wuxi

China

Sunway SW26010 260C 1.45GHz

10,649,600 Cores

1,310,720 GB RAM

125,436 TFlop/s 15,371.00 kW
3 Sierra National Institute of Advanced Industrial Science and Technology (AIST)

Japan

Xeon Gold 6148 20C 2.4GHz, 391,680 cores

417,792 GB RAM

32,576.6 TFlop/s 1,649.25 kW


4 Tianhe-2 (MilkyWay-2)

 

National Super Computer Center in Guangzhou

China

Intel Xeon E5-2692 12C 2.200GHz, TH Express-2, Intel Xeon Phi 31S1P

2,277,376 GB RAM

4,981,760 Cores

100,679 TFlop/s 18,482.00 kW
5 AI Bridging Cloud Infrastructure (ABCI) National Institute of Advanced Industrial Science and Technology (AIST)

Japan

Xeon Gold 6148 20C 2.4GHz processor

391,680 Cores

417,792 GB

32,576.6 TFlop/s 1,649.25 kW
6 Piz Daint Cray Inc.
Swiss National Supercomputing Centre (CSCS)Switzerland
Cray XC50, Xeon E5-2690v3 12C 2.6GHz, Aries interconnect , NVIDIA Tesla P100

361,760 Cores

340,480 GB RAM

25,326.3 TFlop/s 2,271.99 kW
7 Titan Cray Inc.
DOE/SC/Oak Ridge National LaboratoryUnited States
Opteron 6274 16C 2.200GHz, Cray Gemini interconnect, NVIDIA K20x

560,640 Cores

710,144 GB RAM

27,112.5 TFlop/s 8,209.00 kW
8 Sequoia IBM
DOE/NNSA/LLNL
United States
BlueGene/Q, Power BQC 16C 1.60 GHz

1,572,864 Cores

1,572,864 GB RAM

20,132.7 TFlop/s 7,890.00 kW



9 Trinity Cray Inc.
DOE/NNSA/LANL/SNL
United States
Cray XC40, Intel Xeon Phi 7250 68C 1.4GHz, Aries interconnect

979,968 Cores

43,902.6 TFlop/s 3,843.58 kW
10 Cori Cray Inc.
DOE/SC/LBNL/NERSC
United States
Cray XC40, Intel Xeon Phi 7250 68C 1.4GHz, Aries interconnect

622,336 Cores

878,592 GB RAM

27,880.7 TFlop/s 3,939.00 kW

1. Summit

The $200 million supercomputer is an IBM AC922 system utilizing 4,608 compute servers containing two 22-core IBM Power9 processors and six Nvidia Tesla V100 graphics processing unit accelerators each. Summit is also (relatively) energy-efficient, drawing just 13 megawatts of power, compared to the 15 megawatts TaihuLight pulls in. Summit boasts a peak performance of 200,000 trillion calculations per second and will be eight times more powerful than America’s current top-ranked system, Titan. It took four years to build, according to Romett

2. Sunway TaihuLight

Currently ranked as the fastest supercomputer in the world, the Sunway TaihuLight supercomputer measures in at 125 petaFLOPS (theoretical peak)—five times as fast as the supercomputer in second place. Housed in the National Supercomputing Center in Wuxi, it is comprised of 10.6 million cores and is being used for climate research, earth systems modeling and data analytics. On top of being the fastest supercomputer in the world, the Sunway TaihuLight is currently ranked as the fourth most energy-efficient one as well, requiring substantially fewer megawatts per megaFLOPS.

3. Sierra

Sierra, LLNL’s next-generation supercomputer, is projected to provide four to six times the sustained performance and be at least seven times more powerful than Sequoia, with a 125 petaFLOP/s peak  based on IBM’s Power9 chip

4. Tianhe-2 (MilkyWay-2)

In late 2016, the Piz Daint supercomputer in Lugano, Switzerland gained a huge hardware upgrade. That new power tripled its computing performance and brought its theoretical peak performance up to 19.6 petaFLOPS (their own measurementspin it currently at 25.3), making it the fastest supercomputer outside Asia. Named after a mountain in the Swiss Alps, the Piz Daint also creates advanced visualizations and high-resolution imaging simulations. It will soon provide processing power to the Large Hadron Collider at CERN, helping it analyze huge amounts of data.

5. AI Bridging Cloud Infrastructure (ABCI)

AI Bridging Cloud Infrastructure (ABCI) is a 130-petaflops supercomputer costing $173 million (£131 million). Japan’s new machine will be used in the field of Artificial Intelligence, which explains its rather boring name: “AI Bridging Cloud Infrastructure,” or ABCI. According to Satoshi Sekiguchi, a director-general at Japan’s ‎National Institute of Advanced Industrial Science and Technology, where the computer will be built, the system will also be used to “tap medical records to develop new services and applications.”

Specifications

  • Manufacturer: Fujitsu
  • Cores: 391,680
  • Memory: 417,792 GB
  • Processor: Xeon Gold 6148 20C 2.4GHz
  • Linpack Performance (Rmax): 19,880 TFlop/s
  • Theoretical Peak (Rpeak): 32,576.6 TFlop/s
  • Power: 1,649.25 kW (Submitted)
  • Measured Cores: 10,880
  • Operating System: Linux

6. Piz Daint

 

Named after Piz Daint, a prominent peak in Grisons that overlooks the Fuorn pass, this supercomputer is a hybrid Cray XC40/XC50 system and is the flagship system for national HPC Service.

  • Piz Daint has been updated in Fall 2016 and incorporates the previous Cray XC30 named Piz Daint & Piz Dora.
    XC50 Compute Nodes: Intel® Xeon® E5-2690 v3 @ 2.60GHz (12 cores, 64GB RAM) and NVIDIA® Tesla® P100 16GB  total of 5320 Nodes.
  • XC40 Compute Nodes Two Intel® Xeon® E5-2695 v4 @ 2.10GHz (2 x 18 cores, 64/128 GB RAM) – 1813 Nodes.
  • Login Nodes:  Intel® Xeon® CPU E5-2650 v3 @ 2.30GHz (10 cores, 256 GB RAM)
  • Scratch capacity 6.2 PB

 

7. Titan

Perhaps one of the best-known supercomputers in the Western world, Titan at Tennessee’s Oak Ridge National Laboratory was the fastest supercomputer on the planet until the Tianhe-2 (below) jockeyed it out of first place in 2013. Titan is the first supercomputer to combine AMD Opteron CPUs and NVIDIA Tesla GPUs, bringing its total theoretical peak output to 27 petaFLOPS (Linpack approximates its output at 17.6). This kind of power enables researchers to perform the complex simulations needed in climate science, astrophysics, and molecular physics.

8. Sequoia

The Sequoia is a supercomputer built to measure the risks of nuclear warfare by making advanced weapons science calculations. It’s owned by the Lawrence Livermore National Laboratory in California. With 98,304 nodes, it’s ranked as the fifth most powerful supercomputer on the planet. According to the Linpack benchmark, it has a speed of 17.2 petaFLOPS.

9. Trinity

Trinity, a Cray XC40 supercomputer at the Laboratory, was recently upgraded with Intel “Knights Landing” Xeon Phi processors, which propelled it from 8.10 petaflops six months ago to 14.14 petaflops.

The Trinity project is managed and operated by Los Alamos National Laboratory and Sandia National Laboratories under the Alliance for Computing at Extreme Scale (ACES) partnership. The system is located at the Nicholas Metropolis Center for Modeling and Simulation at Los Alamos and covers approximately 5,200 square feet of floor space.

10. Cori (NERSC)

The National Energy Research Scientific Computing Center near Oakland, California named its newest supercomputer creation “Cori,” after Gerty Cori, the first American woman to win a Nobel Prize. The system is a Cray XC40, manufactured by the company responsible for major breakthroughs in supercomputer performance during the 1970s. Cori can theoretically achieve a processing speed of 29.1 petaFLOPS. It achieves this through the use of Haswell architecture Intel Xeon and Xeon Phi processors.

 

 

5.0
OVERALL SCORE
Review
Have no any user vote

COMMENTS

DISQUS: 0