SUPERCOMPUTERS

Supercomputers

Written by - Rahul Jha
Uploaded on - 15/08/2022

INTRODUCTION:


A supercomputer is a device with a high level of performance as compared to a general-purpose computer. Its performance is commonly measured in floating-point operations per second (FLOPS) instead of million instructions per second (MIPS). Since November 2017, the fastest 500 supercomputers use Linux-based operating systems. They have certain features which differentiate them from other computers. They generally have more than one central processing unit, which contains circuits for interpreting program instructions and executing arithmetic and logic operations in the right sequence. Another distinguishing feature of supercomputers is their use of vector arithmetic—i.e., they can operate on pairs of lists of numbers rather than on mere pairs of numbers.


HISTORY


In the 1960s, supercomputers were introduced and for several decades the fastest supercomputers were made by Seymour Cray at Control Data Corporation (CDC), Cray Research, and companies bearing his name or monogram. The US has long been the leader in the supercomputer field followed by Japan, which made significant development in the 1980s and 90s. China has made recent developments in this field too.


HOW SUPERCOMPUTER WORKS


The supercomputer consists of many central processing units (CPUs). These CPUs have groups composed of compute nodes and memory. To solve problems, supercomputers can contain thousands of nodes that use parallel processing to communicate with one another. They use vector arithmetic to operate on pairs of lists of numbers rather than on mere pairs of numbers.


USAGE AND DEPLOYMENT


Supercomputers play an important role in the field of computational science. They are used for a wide range of computationally intensive tasks in fields like quantum mechanics, weather forecasting, climate research, oil and gas exploration, molecular modeling (computing the structures and properties of chemical compounds, biological macromolecules, polymers, and crystals), and physical simulations (such as simulations of the early moments of the universe, airplane and spacecraft aerodynamics, the detonation of nuclear weapons, and nuclear fusion). They have been essential in the field of cryptanalysis.


Some of the functions of a supercomputer also find their application in cloud computing.

PERFORMANCE MEASUREMENT


Generally, supercomputers aim for maximum in-capability computing rather than capacity computing. Capability computing is thought of as using the maximum computing power to solve a single large problem in the least amount of time. Often a capability system can solve a problem of a size or complexity that no other computer can. A single number cannot reflect the overall performance of a computer system, the goal of the Lin pack benchmark is to estimate how fast the computer solves numerical problems and it is widely accepted in the industry. It gives some idea of performance for some real-world problems, but does not necessarily match the processing requirements of many other supercomputer workloads, which may require more memory bandwidth, or may require better integer computing performance, or may need a high-performance input/output system to achieve high levels of performance.


FASTEST SUPERCOMPUTERS


  1. Frontier, the new number 1, is built by Hewlett Packard Enterprise (HPE) and installed at the Oak Ridge National Laboratory (ORNL) in Tennessee, USA.
  2. Fugaku, which previously held the top spot, is housed at the Riken Center for Computational Science in Kobe, Japan. It is thrice as fast as the next supercomputer in the top 10.
  3. LUMI is Hewlett Packard Enterprise (HPE) system, that crunches the numbers in Finland.
  4. Summit, developed by an IBM-built supercomputer, is also at ORNL in Tennessee. Summit is used to tackle climate change, predict extreme weather, and understand the genetic factors that influence opioid addiction.
  5. Sierra, another US entry is a system installed at the Lawrence Livermore National Laboratory in California, which is used for testing and maintaining the reliability of nuclear weapons.
  6. China's best performer is the Sunway TaihuLight, it is developed by the National Research Center of Parallel Computer Engineering and Technology and operational in Wuxi, Jiangsu.
  7. Perlmutter is another top 10 entry from HPE technology.
  8. Selene is a supercomputer currently operational at AI multinational NVIDIA in the US.
  9. Tianhe-2A, built by the China's National University of Defense Technology and giving service at the National Supercomputer Center in Guangzhou.
  10. France's Adastra is the second-fastest supercomputer in Europe and has been developed using HPE and AMD technology.


FUTURE OF SUPERCOMPUTERS


The rising cost of operating supercomputers has been a driving factor in an effort toward bundling resources through a distributed supercomputer infrastructure. National supercomputing centers, which first emerged in the US have been set up in Germany and Japan. Financing supercomputer hardware also became increasingly difficult. In the mid-1990s a top 10 supercomputer required about 10 million euros, while in 2010 the top 10 supercomputers needed an investment of between 40 and 50 million euros.


The supercomputer and high-performance computing (HPC) market is growing at a rapid pace as more vendors like Amazon Web Services, Microsoft, and Nvidia develop their supercomputers. HPC is becoming more important as AI capabilities gain traction in all industries from predictive medicine to manufacturing. In 2020, Hyperion Research predicted that the supercomputer market will be worth $46 billion in the next 4 four years.