The Mightiest Computers on Earth

Over the decades, we have evolved, not just as human beings, but also in terms of technology. A simple smart-phone today can do more that what the best personal computer could have done ten or fifteen years ago. And when this thought comes into our minds, we can't help but think "If such a small system can do so much, what are we actually capable of doing with supercomputers?" So let's look into some of the fastest computers on this planet today, May 2017.

The Computers are ranked based on their Floating point Operations Per Second (FLOPS), or in layman terms, calculations per second. However, at this point a supercomputer is capable of certain Quadrillion calculations per second, thus we use Peta FLOPS. The speed pf these systems were calculated using the LINPACK benchmarks.

10. Trinity, Los Alamos National Laboratory

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 capabilities of Trinity are required for supporting the NNSA Stockpile Stewardship program’s certification and assessments to ensure that the nation’s nuclear stockpile is safe, reliable, and secure. It was built at a cost of $174 M by Cray Inc. and runs at 8.1 Peta FLOPS.

9. IBM Mira, Argonne National Laboratory

Argonne scientists used Mira to identify and improve a new mechanism for eliminating friction, which fed into the development of a hybrid material that exhibited super-lubricity at the macroscale for the first time simulating up to 1.2 million atoms for dry environments and up to 10 million atoms for humid environments. It was built by IBM at a cost of $180 M and runs at 8.5 Peta FLOPS.

8. Piz Daint, Swiss National Supercomputing Center

The Piz Daint is used for complex simulations at the CSCS. The CSCS or the Swiss National Supercomputing Center is open to all Swiss researchers and their assistants, who can get free access to CSCS' supercomputers in a competitive scientific evaluation process. The Piz Daint has a 206,720 cores Intel Haswell EP + Nvidia Tesla GPU, running at 9.779 Peta FLOPS.

7. K, Riken

The K computer at thee Riken Advanced Institute for Computational Science, is used for a variety of applications, including climate research, disaster prevention and medical research. The K computer comprises 88,128 2.0 GHz eight-core SPARC64 VIIIfx processors contained in 864 cabinets, for a total of 705,024 cores, manufactured by Fujitsu with 45 nm CMOS technology. It runs at 10 Peta Flops with peak performance of 11.28 Peta FLOPS.

6. Oakforest PACS, Joint Center for Advanced High Performance Computing, Kashiwa

The Oakforest-PACS system is located in the Information Technology Center on the University of Tokyo's Kashiwa Campus, but everything is carried out jointly by the University of Tokyo and the University of Tsukuba, including financing, implementation and operation of the system, as well as the majority of program usage. It is used for Advanced computation research and has a potential for other applications as well. It runs at 13 Peta FLOPS with a theoretical peak of 24 PFLOPS.

5. Cori, National Energy Research Scientific Computing Center

It was customized to support data-intensive science and the analysis of large datasets through a combination of hardware and software configurations and queue policies. Cori runs using Intel Xeon Phi processors at 14 pFLOPS.

4. IBM Sequoia, Lawrence Livermore National Laboratory

Sequoia will be used primarily for nuclear weapons simulation, replacing the current Blue Gene/L and ASC Purple supercomputers at Lawrence Livermore National Laboratory. Sequoia will also be available for scientific purposes such as astronomy, energy, lattice QCD, study of the human genome, and climate change. Sequoia is a Blue Gene/Q design, based on previous Blue Gene designs. It consists of 96 racks containing 98,304 compute nodes, i.e., 1024 per rack. The compute nodes are 16-core A2 processor chips with 16 GB of DDR3 memory each. Thus, the system contains a total of 1,572,864 processor cores with 1.5 PiB memory. It covers an area of about 3,000 square feet (280 m²). The compute nodes are interconnected in a 5-dimensional torus topology. This lets the system run at a massive 20 pFLOPS peak.

3. Titan, Oak Ridge National Laboratory

Titan is available for any scientific purpose; access depends on the importance of the project and its potential to exploit the hybrid architecture. Any selected programs must also be executable on other supercomputers to avoid sole dependence on Titan. Titan employs AMD Opteron CPUs in conjunction with Nvidia Tesla GPUs to improve energy efficiency while providing an order of magnitude increase in computational power over Jaguar. It uses 18,688 CPUs paired with an equal number of GPUs to perform at a theoretical peak of 27 pFLOPS.

2. Tianhe-2, Guangzhou

In Chinese, the "Milky Way - 2", is being mentioned for the second time on this blog. With 16,000 computer nodes, each comprising two Intel Ivy Bridge Xeon processors and three Xeon Phi coprocessor chips, it represented the world's largest installation of Ivy Bridge and Xeon Phi chips, counting a total of 3,120,000 cores. With a speed of 33 pFLOPS, this was the fastest computer on Earth in 2013, however, it has faced a lot of criticism for beinng difficult to use. "Some users would need years or even a decade to write the necessary code" says the deputy director of Computer Network and Information Center. The system was developed for the sole purpose of breaking a speed record and thus, has a limited usability.

1. Sunway TaihuLight, Wuxi

TaihuLight is currently up and running at the National Supercomputing Center in the city of Wuxi, a manufacturing and technology hub, a two-hour drive west of Shanghai. The system will be used for various research and engineering work, in areas such as climate, weather & earth systems modeling, life science research, advanced manufacturing, and data analytics. The Sunway TaihuLight uses a total of 40,960 Chinese-designed SW26010 many-core 64-bit RISC processors based on the Sunway architecture. Each processor chip contains 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. This lets the computer run at a peak of 125 pFLOPS.

Supercomputers are becoming faster and smarter day by day. They calculate our weather forecasts, disaster management, and other analytical models. They can simulate what humans cannot imagine and that's giving us the cutting edge technology of the future.