what is supercomputer

hpc-eni fbg

On hearing the name of Supercomputers, we would come to the image of those huge, very large and overheating machines which are famous for doing the most fast calculating in the world, which took huge amounts of space for computation after computation.

You may be surprised to learn that with the ubiquitous nature of personal PCs and networked systems, supercomputers are still used in various operations.

In the next few minutes, I will tell you about supercomputers and how they still work in many industrial and scientific fields.

What is a supercomputer?

A supercomputer is a computer that is one of the most powerful, fastest systems in the world at any given point in time.

For example, the first supercomputer was in the UK, named Colossus. It was designed to read messages and crack German codes during World War II, and it could read 5,000 characters in a second.

It sounds impressive, doesn't it? Colossus was considered very powerful at the time, but if you compare it to the NASA Columbia supercomputer that completes 42 million operations per second, it sounds very childish.

In other words, the capacitance of what used to be a supercomputer is now less than your mobile, and what we currently call supercomputers is as advanced as other computers can be.

However, there are some things that make the computer branch in the "super" sector. It usually consists of more than one Central Processing Unit (CPU), which allows the computer to perform rapid circuit switching and complete more tasks at once. (Because of this, a supercomputer will also have a large amount of storage so that it can access multiple tasks at a time.) It will also have the ability to perform vector arithmetic, which means that it can list multiple operations at once. Can calculate

Computing the performance of a "supercomputer" is much higher than that of a general-purpose computer. The computing performance of a supercomputer is measured in FLOPS (Floating-Point Operations Per Second) rather than MIPS.

Supercomputers have thousands of processors that can do billions and trillions of calculations per second, or you could say that supercomputers can reach about 400 quadrillion FLOPS.

They have evolved from grids of parallel parallax computing to cluster systems. Cluster system computing means that the machine uses multiple processors in a system instead of arrays of individual computers in the network.

These computers are the largest in size. A most powerful supercomputer can grow from a few feet to hundreds of feet. Supercomputers also cost a lot and can cost from $ 2 million to $ 100 million.

History of Supercomputers

After studying the history of computers, you will understand that no one person can claim to have invented these amazing machines.

But arguably, this is much less true of the supercomputer, which is widely acknowledged to have contributed immensely to the work of the same man, Seymour Cray (1925–1996). Here's a tour before and after Supercomputing, BC and AC!

1946: John Mauchly and J. Prepper Eckert created the ENIAC (Electronic Numerical Integrator And Computer) at the University of Pennsylvania.

It was the first general-purpose electronic computer, which is approximately 25 meters (80 ft) long and weighs 30 tons, and since it is deployed on military-scientific problems, it is arguably the first scientific supercomputer.

1953: IBM developed its first general-purpose mainframe computer, the IBM 701 (also known as the defense calculator).

It sold about 20 machines to various government and military agencies.

The 701 is arguably the first off-the-shelf supercomputer. IBM engineer Jean Amahl later redesigned the machine to create the IBM 704, a machine capable of 5 KFLOPS (5000 FLOPS).

1956: IBM developed the Stretch supercomputer for the Los Alamos National Laboratory. It remained the fastest computer in the world until 1964.

1957: Seymour Cray, co-founder of Control Data Corporation (CDC) and Pioneer Fast, transistorized fast, high-performance computers, including CDC 1604 (announced 1958) and 6600 (released 1964). This challenged IBM's dominance in mainframe computing.

1972: Cray abandons control data and founded Cray Research to develop high-end computers - the first true supercomputer. One of their main ideas is to help reduce the length of connections between components inside their machines so that they can be built faster. This is partly the reason why early Cray computers were C-shaped, although the unusual surface design (and bright blue or red cabinet) also helps distinguish them from competitors.

1976: The first Cray-1 supercomputer is installed at the Los Alamos National Laboratory.

It manages a speed of around 160 MFLOPS.

1979: Cray developed another fast model, eight processors, 1.9 GFLOP Cray-2. Where wire connections in the Cray-1 were a maximum of 120 cm (~ 4 ft) long, in the Cray-2 they grew to only 41 cm (16 in).

1983: Thinking Machines Corporation unveils a large-scale parallel connection machine with 64,000 parallel processors.

1989: Seymour Cray started a new company, Cray Computer, where they developed Cray-3 and Cray-4.

1990: Defense spending cuts and the rise of powerful RISC workstations created by companies such as Silicon Graphics pose a serious threat to the financial viability of supercomputer manufacturers.

1993: Fujitsu Numerical Wind Tunnel became the world's fastest computer using 166 vector processors.

1994: Thinking Machines for Bankruptcy Protection

1995: Cray Computer runs into financial difficulties and filed for bankruptcy protection. Sadly, Seymour Cray died on October 5, 1996, due to injuries in a car accident.

1996: Purchased by Cray Research (Cray's parent company) Silicon Graphics.

1997: ASCI Red, a supercomputer made of Pentium processors by Intel and Sandia National Laboratories, has become the world's first TeraFLOP (TFLOP) supercomputer.

1997: IBM's Deep Blue Supercomputer defeated Gary Kasparov in chess.

2008: The Jaguar Supercomputer, built by Cray Research and Oak Ridge National Laboratory, became the world's first Petaflop (PFLOP) scientific supercomputer. In short, the world's fastest computer, it was soon superseded by machines in Japan and China.

2011–2013: Jaguar was extensively upgraded, named Titan, and became the world's fastest supercomputer before the Chinese machine Tianhe-2 topped.

2014: Mont-Blanc, an EU, announced plans to build ExaFLOP (1018 FLOP) supercomputers from energy-efficient smartphone and tablet processors.

2017: Chinese scientists announced that they would soon unveil a prototype of an Xflop supercomputer, expected to be based on the Tianhe-2.

2018: China now dominates the TOP500 rankings of the world's 500 fastest supercomputers, surpassing the United States by 202 machines to 143 (a year earlier, both countries claimed 171 machines). In June 2018, the 200-petaflop supercomputer reclaimed the number one spot for the United States at Oak Ridge's new summit.

Supercomputer Features

1. They can support more than a hundred users at a time.
2. These machines are capable of handling large amounts of calculations beyond human capabilities, that is, humans are unable to solve such extensive calculations.
3. Many people can access a supercomputer at the same time.
4. These are the most expensive computers that can ever be built.

How powerful are supercomputers?

Look at the specifications of ordinary computers and you will find that their performance is usually measured in MIPS (Million Instructions Per Second), which can manage how many fundamental programming commands (read, write, store and etc.).

It is easy to compare two PCs by comparing MIPS numbers (or even their processor speed, usually rated at GHz or GHz).

Supercomputers are rated in a different way. Since they are employed in scientific calculations, how many floating-point operations they can perform per second, which they can do, which is exactly what they are trying to do, which, unlike MIPS, is a more meaningful measurement. It is a measure of how they are trying to do it). Since supercomputers were first developed, their performance has been consistently measured in large numbers of FLOPS

What do supercomputers actually do?

As we saw at the beginning of this article, an essential feature of a computer is that it is a general-purpose machine that you can use in various ways: you can send email to a computer, play games, photos You can edit or run a program and just do some other things.

If you are using an Android phone or a high-end cellphone such as an iPhone or iPod touch, you have a powerful small pocket computer that can load various "apps" (applications) and run programs, which are simply computer programs, But by any other name. Supercomputers are slightly different.

Typically, supercomputers are used for complex, mathematically intensive scientific problems, including simulating nuclear missile tests, weather forecasting, simulating the climate, and testing the strength of encryption (computer security code). In theory, a general-purpose supercomputer can be used for absolutely anything.

While some supercomputers are general-purpose machines that can be used for a wide variety of different scientific problems, some are engineered to perform very specific jobs. Two of the most famous supercomputers in recent times were engineered in this way. From 1997 IBM's Deep Blue Machine was created specifically to play chess (against Russian Grand Master Gary Kasparov), while later the Watson machine (named after IBM's founder, Thomas Watson, and his son) played the game Jeopardy. Was engineered for.

Such specially designed machines can be customized for special problems; So, for example, Deep Blue was designed to search through a vast database of potential chess moves and evaluate which moves might be best in a particular situation, while Watson was given the (natural) It was customized to analyze difficult general knowledge questions described in human language.

By modeling the Earth's Tri-D structure in other supercomputer simulations, researchers can predict how earthquake waves will travel locally and globally.