Overview of Supercomputers

 

Supercomputers

What is Supercomputer?

A supercomputer is a computer that performs at a higher level than a general-purpose computer. A supercomputer's performance is measured in floating-point operations per second (FLOPS) rather than million instructions per second (MIPS). Supercomputers are massive computers with tens of thousands of processors that can do billions or trillions of calculations per second. Some supercomputers have the ability to perform up to 100 quadrillion FLOPS. Supercomputers are useful for real-time applications because information travels quickly between processors in comparison to dispersed computing systems.

Supercomputers are used for Quantum physics, weather forecasting, oil and gas exploration, molecular modelling, physical simulations, aerodynamics, nuclear fusion research, and cryptoanalysis are just some of the data-intensive and computation-intensive scientific and engineering applications. To boost the speed of early supercomputers, custom made operating systems were created for each one. Supercomputer design has shifted away from proprietary, in-house operating systems and toward Linux in recent years. Despite the fact that most supercomputers run on Linux, each manufacturer optimizes its own Linux variant for maximum hardware performance. SUSE Enterprise Linux Server was utilized by half of the world's top 50 supercomputers in 2017.

 

Differences between general-purpose computers and supercomputers

Supercomputers are general-purpose computers that operate at the fastest possible rate or provide the best possible performance. The major difference between supercomputers and general-purpose computer systems is processing power. A supercomputer can perform 100 PFLOPS. A typical general-purpose computer can only do a few hundred gigaflops to tens of teraflops of computation.

Supercomputers use a lot of energy. As a result, they generate so much heat that cooling systems are required to store it.

Quantum computers, which work on quantum physics principles, are different from both supercomputers and general-purpose computers.

What are supercomputers used for?

Supercomputers are used to execute resource-intensive calculations that are beyond the capabilities of general-purpose computers. Engineering and computational sciences applications, such as the following, are frequently run on them:

Data Mining is used to extract information from raw data gathered from data farms on the ground or in the cloud.

 

Intelligence Agencies are used to monitor communications between private citizens, or from suspected terrorist organizations and potentially hostile governments.

 

Cryptanalysis to analyze, cyphertext, ciphers and cryptosystems to understand how they work and identify ways of defeating them.

 

Oil and gas exploration to collect huge quantities of geophysical seismic data to aid in finding and developing oil reserves;

 

Molecular modeling for calculating and analyzing the structures and properties of chemical compounds and crystals;

 

Physical simulations like modeling supernovas and the birth of the universe;

 

Aerodynamics such as designing a car with the lowest air drag coefficient;

 

Nuclear fusion research to build a nuclear fusion reactor that derives energy from plasma reactions;

 

Medical research to develop new cancer drugs, understand the genetic factors that contribute to opioid addiction and find treatments for COVID-19;

 

Next-gen materials identification to find new materials for manufacturing; and

 

Weather forecasting to predict the impact of extreme storms and floods;

Supercomputers, like any other computer, are used to replicate reality on a bigger scale. Cloud computing may perform some of the same functions as a supercomputer. Cloud computing, like supercomputers, uses the processing power of several processors to achieve performance that is unattainable on a PC.

When was the supercomputer first invented?

The word "supercomputer" initially appeared in the early 1960s, when IBM released the IBM 7030 Stretch and Sperry Rand released the UNIVAC LARC, the first two purposeful supercomputers meant to outperform the fastest commercial machines at the time. The US government began sponsoring the development of cutting-edge, high-performance computer technology for military uses in the late 1950s, which affected the rise of supercomputing.

Supercomputing in India

Supercomputing in India started in 1980 when the Indian government installed an indigenous improvement program as there had been several troubles you bought supercomputers from abroad. The countrywide Aerospace Laboratories started the challenge “Flosolver MK1”, a parallel processing gadget operating in December 1986. Following this, a couple of projects had been commissioned from different organizations, which include C-DAC, C-DOT, NAL, BARC, and ANURAG. C-DOT created “CHIPPS”, the C-DOT excessive-performance Parallel Processing gadget, and BARC created the Anupam collection of supercomputers. ANURAG created the tempo series of supercomputers.

Even though the C-DAC task released the “PARAM” series of the supercomputer, it was most effective in 2015 that the release of the national tremendous Computing challenge boosted the Indian supercomputers. NSM introduced a seven-year program well worth Rs four,500 crores to install seventy-three indigenous supercomputers through 2022. As of November 2020, PARAM Siddhi-AI is the fastest supercomputer in India and 63rd inside the TOP500 listing.

 

Supercomputers under C-DAC mission:

In November 1987, the Centre for improvement of Advanced Computing (C-DAC) was allotted a 3–12 months finances of Rs 375 million to create 1000Mflops (1Gflops) supercomputers. The “PARAM” (brief shape for Parallel device) collection of supercomputers was unveiled in three C-DAC missions.

• PARAM 9000
• PARAM 10000
• PARAM Padma
• PARAM Yuva
• Param Yuva II
• PARAM ISHAN
• PARAM Brahma
• PARAM Siddhi-AI

Supercomputers under the National Supercomputing Mission :

• PARAM Shivay
• PARAM Sanganak
• PARAM Pravega

Supercomputers and artificial intelligence

Supercomputers and artificial intelligence

Artificial intelligence (AI) applications are frequently run on supercomputers since they typically require supercomputing-level performance and processing power. The large amounts of data used in AI and machine learning application development can be handled by supercomputers.

Some supercomputers are specifically designed with AI in mind. Microsoft, for example, constructed a supercomputer especially for training massive AI models that interact with its Azure cloud platform. The goal is to employ Azure’s AI services to give supercomputing resources to developers, data scientists, and business customers. Turing Natural Language Generation, a natural language processing model from Microsoft, is one such technology.

Nvidia’s Perlmutter is another example of a supercomputer designed exclusively for AI tasks. It is ranked №5 among the world’s fastest supercomputers in the most recent TOP500 list. It will be tasked with creating the world’s largest 3D map of the visible universe, with 6,144 GPUs. It will achieve this by analyzing data from the Dark Energy Spectroscopic Instrument, a camera that takes dozens of images per night of thousands of galaxies. 

The market for supercomputers and high-performance computing (HPC) is expanding as more companies construct their own supercomputers, such as Amazon Web Services, Microsoft, and Nvidia. As AI skills gain popularity in a variety of industries, from predictive medicine to manufacturing, HPC is becoming more crucial. According to Hyperion Research, the supercomputer market will be valued at $4 billion in 2024.

The race to exascale processing capability is the current focus in the supercomputer market. Exascale computing has the potential to open up new possibilities beyond even the most advanced supercomputers. Exascale supercomputers should be able to create a realistic model of the human brain, complete with neurons and synapses. The field of neuromorphic computing would be forever changed as a result of this. Supercomputers with hundreds of exaflops may become a reality as computing power continues to rise exponentially.

As AI plays a larger role in enterprise computing, supercomputers are becoming more common.

Conclusion

We now live in an era of information technology, which would not be possible without computers. Ordinary computers, often referred to as general-purpose computers, are the computers that we use on a daily basis. We are capable of performing both easy and challenging tasks with ease. Simple occupations include things like doing simple math, looking at a picture, and typing text. Large and high-performance projects, on the other hand, necessitate the use of supercomputers. We learn what a supercomputer is and how it functions. What’s the use of having a supercomputer? When was the first supercomputer created? The distinctions between general-purpose and supercomputer computers.

References:

1. https://www.youngwonks.com/blog/What-is-a-Supercomputer-and-What-are-its-Types--Uses-and-Applications
2. https://blog.datawrapper.de/the-race-to-build-the-fastest-supercomputer/
3. https://corporate-blog.global.fujitsu.com/fgb/2021-02-15/how-supercomputer-technology-is-improving-everyday-life/
4. https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.infoworld.com%2Farticle%2F3642848%2Fthe-cloud-as-supercomputer.html&psig=AOvVaw3Ao_MXpZQ2wLYLm_aJebyr&ust=1654177982772000&source=images&cd=vfe&ved=0CAwQjRxqFwoTCMDf862zjPgCFQAAAAAdAAAAABAD
5. https://www.google.com/imgres?imgurl=https%3A%2F%2F6lli539m39y3hpkelqsm3c2fg-wpengine.netdna-ssl.com%2Fwp-content%2Fuploads%2F2021%2F06%2F2021_059_Supercomputer-des-KIT-einer-der-15-schnellsten-in-

Authors:

1. Shruti Pattewar
2. Kuhu Mukhopadhyay
3. Tanmay Mutalik
4. Aishwarya Patil
5. Rehanuddin Qazi