After half a century back, even the smallest computer at that time occupied a whole room. And today, with the use of microchip, more powerful computers have been pocketed in the past. But have you ever wondered, if these small chips are made of a house-sized computer then how strong will it be? Is this possible to create a super computer? Will get a million times more than your laptop or desktop to get supercomputing, and can solve any complex scientific problem in the world? Today’s question is, what type of feature helps to make a simple computer super-computer? How different is a super-computer from your current computer? Let’s learn all the information about supercomputing.
What is a supercomputer?
Computers can be extremely fast, solve complex scientific problems, or predict the weather, or just say what the climate of 2050 will be like and what such a great work can be done as a supercomputer. Now you might ask, how fast the computer to work? If my computer has more RAM and more core processors, more GPUs, but can I do supercomputing? In order to fully understand the definition given by the supercomputer in “simple definition”, you have to be aware of some terms first.
This is basically a machine that completes any normal work. It takes an input to the first input and then processing them on your own, and eventually, an output result appears in front of you. In fact, supercomputer means that there must be a huge giant computer or your computer must be multi-billion times faster, but this is not the case, actually, it uses a completely different method to complete any work. Like your normal desktop or laptop, it does a lot of work at a time, not just doing one thing at a time.
And this is basically the main feature that makes a computer a supercomputer.
Serial and parallel processing
What is the difference between serial and parallel processing? A simple computer can do one thing at a time, ie, to complete a task, it provides an output of the job by completing one after another process. And it’s called serial processing. Now you will say, “I play music with my computer at the same time, browse the internet and render the video, and you say one thing at a time”? In fact, your computer can complete a few million orders per second; in this case, you have a real-time serial no matter what you do. Let’s try to make the subject more clearly by a nice example. Imagine a simple shopkeeper, think you went to his shop and asked him to pack a kilogram of dough.
Now what he will do – he will first get up from his seat, then go to the rack to keep the packet of flour, bring the flour packet, hand it over to you, and calculate your given cash and keep it in the drawer. Now, see how fast he got up from his seat, or how quickly he took flour from the rack and handed it to you. It is not the main issue. The main thing is that he is only doing one job at a particular time. No matter how quickly he supplies you, he is doing one thing at a time and providing service to you by doing one after another. In the same way normal computers complete the work.
But today’s modern supercomputers work using a completely different method. It divides the work into several parts, and simultaneously processes them, and this system is called Parallel Processing. Now, as per the above example, you went to a shop to buy 1 kg flour, 1 liter of oil, and 1 kg of sugar. Now suppose that there are 3 people in the shop to help the shopkeeper. Then all you can do is pick a different job to supply the things you want. Someone will bring sugar, someone will bring oil and someone will supply flour and in the same time you can also remove the money transaction with the shopkeeper. Again, each one brings different things at the same time, but all are coming together to you. In this way, if you do not want to buy anything in the shop, there are many people working, but it is possible to complete all the tasks very fast in the same time. And this is parallel processing; theoretically, our brain also works in the parallel process. At the same time you are reading this article, and at the same time, your brain continues to insinuate the internal activities of your body.
Why does Parallel Processing need for Super Computing?
The computer used by our daily work does not have to be so efficient. To access internet browsing, email, or type your computer has to spend very little processing power. But if you want to do a lot of complex tasks such as changing the color of a large digital photo, your computer will have to do a lot of processing, sometimes it may take up to a few minutes to complete it. Again, if you want to do gaming without internet browsing, email, or typing on your computer, you will need a lot of fast processors, dedicated GPUs and more memory (RAM), or your system will work slow. Fast processor, and twice the amount of memory your computer will start to work very dramatically. But how fast? There is also a limit to this speed-a processor can only work only once.
Now think you are a scientist and you work in a weather office. You have to work together to forecast weather forecasts, new cancer medication tests, and climate models until 2050. Now you can upgrade your PC processor and more memory to complete this task, even though it will work faster than before but there is a limit to its work. But if your work is divided into one piece and each piece is worked with different processors, that means parallel processing, but it will be possible to do your work faster and efficiently than before.
Supercomputers need super-computing to work together a lot of processors – so that they can work together in parallel to accomplish many tasks at the same time. The fastest supercomputer name of the current time is The Sunway TaihuLight. There are 64-bit 40,960 multi-processor processors, and each processor chip has 256 processing cores. That is, this whole computer has 10,649,600 processors!
Parallel computing is a standard system for supercomputing, but parallel computing has some serious problems. Let’s go back to the shop example, but this time I’ll take an example of the super market. Think you went with some of your friends to the supermarket to buy some things. Now if you send your friends to different counters to take some things, your things will be bought in less time, and at the end of the shopping, all things will be gathered to you. In such a way, the supercomputer trims the work and complete the processors and then collect those pieces together to give output. But if any work is too big, sending the pieces to the processors, and after the completion of the work, it becomes a very difficult task for computers to combine all the pieces.
To do this, another management system needs to be done so that each piece is broken into the processor to the processor and the pieces returned are given together and outputs. But to process this management system, there is a lot of processing power needed, which can overload the system. As the task manager of Windows OS-it mainly displays the processing of all programs, but it also requires some processing power to display them. You can complete your shopping in fewer times by sending your friends to different counters, but it will be late for the payment.
You can complete your shopping in fewer times by sending your friends to different counters, but it will be late for the payment.
Think you’ve come to shop millions of things together and 50,000 friends are collecting things from different counters.
If you were shopping at a counter at a time, the cash pay problem would have ended in a counter, but due to the shopping in different counters, a problem has been broken into 50,000 problems.
Think similarly, you command the computer to predict the weather for the week ahead of the whole world. Now your computer shreds all these processes into different processors, to process them if you did the process, then all the processing would have ended at once. But each processor will process different weather forecasting in different countries in order to process different processors. But the problem may be more serious because the weather in different countries of the world can be completely different from other countries. As a result, a processor will have to wait until another processor’s work is finished-and one needs to get help from another processor to solve the problem. So super parallel work on some computers and some tasks need to be done in serial processing.
Types of Supercomputing
It is possible to do a lot of complex tasks by installing parallel processing by creating thousands of processors, RAMs, and GPUs in a huge box and creating supercomputers. Or if you want to be able to interconnect with a lot of PCs in one room with many fast LANs (local area network), you can create a super computer. This type of supercomputer is called Cluster. Google uses this type of computer to provide search results of its users.
Another kind of super-computer name is Grid – which is very similar to clusters, but different computers are located in different parts of the globe and are connected to each other using the Internet or another computer network. This is an example of a distributed computing, where computers in different parts of the world work together on a network. It can say virtual supercomputing.
Different universities or research centers in the world connect their computers with each other and create a grid super computer.
Not every computer in the grid works together, or they do not have to work, but they create a strong computing system together.
Operating of Supercomputer
Cluster super computers are put together in a big place, sometimes it can be equal to 3-4 football fields. And this giant size computer requires a huge amount of electricity to run; this amount is enough to run thousands of homes by electricity. Millions of dollars’ worth of dollars needed to handle super computers for so much electricity costs. Again, supercomputers make a lot of heat, when the electricity flows through the cable when the electricity flows through a cable, it also produces heat energy. And for this, your normal computer also needs a cooling system. But these huge biggies cannot be cooled by ordinary fans. Today’s modern supercomputers are cooled by fluids, exactly the way the fridge works. It costs a lot of electricity to handle this cooling system; secondly, it is also a hammer for the environment.
As already mentioned, a computer is a machine that can do all normal work (essentially virtual). If you want you can run on your computer using your computer’s operating system, like Windows! But most supercomputers run on a Linux-based operating system.
These operating systems are highly customized, meaning they are only made with certain work features. Since supercomputers are used mainly to solve various scientific problems, what do they need to have other work features?
Why Need Supercomputing
As a computer is a machine for all the normal work, you can read the article on internet browsing, email send, photo editing, video editing, and gaming and even you can read techforlive.com article with super-computers. You will need to run different programs on your system to perform this task just like your normal computer.
For example, you need different apps (apps) to do different things on your Android or iPhone, but these are computer programs, but what else is different! These computers are not working on supercomputers for the time being. It is used to solve complex mathematical problems, solve scientific problems, test nuclear intelligence, weather forecasts, climate change, and intensification of intensities. However, theoretically, all the work done by a supercomputer is possible, which is your normal computer.
How powerful is the super computer?
Generic computers publish their ability to work in MIPS or Million Instructions per Second. It indicates that the system can perform various commands (read, write, data store) processing to complete any work. It is easy to compare between two processors via MIPSs, which is said to be as strong as the processor can work at MIPS for as long as possible. This is usually known as processor speed, which is usually expressed as gigahertz. But the power of working on supercomputers is expressed separately. Since these computers are used for scientific purposes, its power is calculated as the Floating Point Operations per Second. This is an efficient method to calculate the effective power of a computer. Let’s try to understand these two terms through a small example.
Imagine, you are watching a football match, you see the player presently 1 ball has the ball, he passed the ball to the player 2 in the field, the 2nd player passed the ball to the player 3 near the goal-and the player 3 beat an intelligent strong He hit the ball and hit the goal. And this is how many other goals are scored through different players’ hits.
Now, here MIPS means that the ball has to hit the ball to reach a goal and hit the ball from the player. And the FLOPS means that the team has scored the total. That is, how many million commands to be followed by a processor to perform a task per second, it is published by MIPS and how many tasks are completed every second; it is published by the FLOPs.