Chinese entertainment for engineering students

Chapter 132 Build a “supercomputer”

"Er Zi, why didn't you tear off the tape? What happened?" Dadong looked curiously at Wang Xu, who was leisurely sunbathing by the window.

"Well, it's done. The results have been accepted. Just wait for the results. How long has it been since you paid attention to me? It was a week ago that I stopped tearing off the tape. Tell me, if you have anything to do, stick your butt out. "Wang Xu curled his lips, Dadong must have something going on if he is so speechless.

"Stop, stop, we're eating, watch your words." Zhou Feng immediately stopped Wang Xu's nonsense.

"I just wanted to ask you if there are any projects that I can do. Isn't this my sophomore year? The professional courses have started, and optics is boring. I see that you have been tossing about this and that all day long, and in the end, you are in the cheap material department. Can you We can't come up with some projects for the class." Dadong was the league secretary of the class last year, but he has resigned this year, but he is still very enthusiastic about class affairs.

By the way, Mizuki's habit is that the league secretary of each class changes every year, and the student union is also very mobile, so it is rare for one person to work all the time. After all, they are all geniuses from all over the world. No one is better than the other. If you don't convince others, you have to balance it.

"Don't tell me, it really is." Wang Xu suddenly thought of an idea.

"Let's talk about it." When several other dormitory brothers heard that there was a project, they all came over and stood or sat in a circle with their lunch boxes.

What, why do you go back to the dormitory to eat instead of eating in the cafeteria and then come back? Well, because there are no seats at all in the Jiu Canteen right now. People can only eat standing up, and there is still no tabletop to put the lunch box on. It is better to go back to the dormitory to eat.

Shuimu has opened several new canteens in the past few years, and the serial number has reached sixteen. The windows of the new canteen are bright and clean, but they are far away, and lazy people are not willing to run away.

"Let's make a supercomputer for fun. Although this thing is more related to the electronics and computer departments. But who are we? Automation is related to everyone, so it's okay to do it." Wang Xu said.

"Pfft~", okay, these brothers sprayed rice, sprayed water, and suddenly the table was in a mess.

"As for that, it's disgusting." Wang Xu complained about the unreliability of these roommates while wiping the rice particles sprayed on his body.

"You are really cool, and you really dare to say anything." Dadong was also wiping the sprayed water, and said while wiping it.

"Supercomputing, the state has invested over 100 million yuan and it hasn't been developed yet. As soon as you talk about it, you are planning to do supercomputing. Although you have made some inventions, don't be so outrageous, okay? You didn't do it before. I know, you can really brag." Boss Zhan Wenbo obviously also thought Wang Xu was talking nonsense.

Indeed, since the Galaxy billion machine in 1983, we have not heard about the progress of domestic supercomputers for many years. The word supercomputer was translated from foreign news in recent years. It was always called supercomputer before.

Supercomputing is naturally one of them, and it is a particularly clear field. When Wang Xu thought of doing this, the main reason was that the current computers were simply not enough.

But today’s supercomputers, not to mention foreign ones are ridiculously expensive, and domestic ones are simply unusable.

For example, Galaxy, with 100 million floating-point operations, is an important scientific research product that breaks the monopoly in China. But in fact, everything inside is soldered by hand. Can you imagine a computer with 2 million solder joints and 400,000 flying wires, all soldered by hand?

This thing is actually to do things that should be done in the CPU directly using circuits. Just like in later generations, Wang Xu had a junior brother who soldered an 8086 CPU directly on a table with thousands of transistors as his graduation project. Galaxy is actually a similar thing.

Not only is there no possibility of mass production and mass manufacturing of this kind of thing, it even has little value in use, because its maintenance is extremely difficult. If there is a problem with any flying wire or solder joint, it will take immeasurable time to troubleshoot.

Therefore, the most important role of Galaxy is not to use it to calculate anything, but to tell the world that we have it, and then the international community will lower some conditions for purchasing supercomputers and break some blockades.

Didn't you realize that as soon as the Galaxy came out, several giant computers were successfully introduced into the country? Although there are still many restrictions including that they can only be used for specific purposes, and foreign personnel must be present to use them, at least they can be purchased.

Moreover, the current supercomputer cannot perform the operations that Wang Xu needs. It does not have a general operating system or commercial software, and all calculations are completed by existing programs.

Therefore, Wang Xu came up with the idea of ​​​​creating a simple version of supercomputer. But this thing cannot be made by Wang Xu alone. Although there are technologies for sale in the system space, after all, computer special effects and VR and other things account for an important proportion of entertainment.

But Wang Xu can't afford these things for the time being. Although they are not expensive, they still have to save for a while. On the other hand, they cannot be made even if they are bought. The industrial fields involved are too many and complex, so he can only draw pictures slowly. Of.

But this doesn't mean that Wang Xu can't do anything now. If he can't do it alone, then he can do it with a group of people. What Shuimu lacks most is talent. What is the purpose of Wang Xu's desperate efforts to gain reputation? Isn't it to make full use of Shuimu's resources? What sheep has more hair and longer than the one in front of him?

Today's personal computers have entered the 32-bit era, but the capabilities of a single computer are still too weak, and minicomputers are too expensive. CPU upgrades are further restricted by Moore's Law and cannot be doubled so quickly.

The current mainstream supercomputer is still a vector computing model like Cray, which is a type of parallel computing. To put it bluntly, this is the computing model of later generations of graphics cards. In later generations, the GPU of the graphics card will simply be used to make Cray machines. , such as the Titan machine in Oak Ridge.

You can think of the core of the current Cray Machine as an oversized GPU made up of a bunch of RISC chips. These chips are the cores in the GPU.

After that, some input, output, storage and other devices are added. Speaking of which, the performance is really not as fast as a later graphics card.

We can almost say that the technology of Cray Machine has been miniaturized and integrated, and it will become the basic GPU.

And although this thing is fast, it is very expensive, super expensive, and consumes a lot of electricity. It is not suitable for small and medium-sized enterprises at all.

So how can we make a computer with simple raw materials and high computing speed?

In fact, future generations will have the answer, and that is distributed computing.

Distributed is a kind of system that uses a special operating system to send divisible computing steps to different computers for calculation, and then summarizes them to obtain the calculation results. This thing is actually the predecessor of the cloud that everyone will be familiar with in later generations.

Parallelism is actually the same as distributed in basic principles, except that it is not allocated to external computers, but to different internal CPUs, so it is faster, but the design is more complex, and the number of CPUs that can be integrated is also limited.

However, talking about speed aside from the application is just being a rogue. First, you have to figure out what you want to do.

If you want to be efficient in parallel algorithms, first of all, what you want to do must support parallel operations, that is, the original operation needs to be broken down into a large number of unrelated simple operations.

For example, 1+2+3+4, you can simply split it into two unrelated operations, 1+2 and 3+4, and then calculate the sum of the results. The operation on a computer requires 3 clock cycles. , and when computing on a parallel computer with two CPUs, only two are needed, which increases the speed.

But some things cannot be split in this way because subsequent operations require the results of the previous step, so parallel operations cannot significantly improve efficiency. For example, 1+2*3, no matter what kind of machine it is on, it takes two clock cycles to complete the calculation.

Distribution also has similar problems. If your work can be divided into many similar subtasks, then it is ideal to use distribution. For example, animation rendering, you can send the material to many computers to render each frame separately.

After the rendering results are obtained, they are first synthesized, then divided into segments, and sent to multiple machines for video generation. Repeat this to get a complete rendering animation.

Audio processing is actually similar to video processing. Audio can also be split into segments and processed separately. Different audio tracks are inherently different subtasks, so audio processing can actually be used. Processed in a distributed manner.

However, audio processing is much simpler than video, and there is no need to split it up so troublesomely. A DSP in later generations can handle it directly.

But today's various CPUs and DSPs are not that powerful yet, so it is a little more troublesome and requires the use of some distributed or parallel technologies.

A high-performance PC is used as the host computer, responsible for human-computer interaction and task distribution, and multiple sets of computers equipped with multiple sound cards are used as computing units to process signals.

Then equip it with a centralized storage unit, use the RAID method to store the transit data, use it as a cache, and use multiple tape drives as permanent storage devices. Well, the general structure is like this.

Does it look familiar? Yes, many small workstations for three-dimensional image rendering in later generations were based on this model, but the sound processing unit was replaced by a graphics card.

In general, building such a system now is actually more costly than gain, because the performance improvement caused by various restrictions will be extremely limited. Moreover, this thing already exists abroad and is quite developed. Otherwise, where do you think the equipment of the top recording studios abroad comes from?

But the problem is, this thing's signal processing capabilities are too strong, and it can be applied to, well, you know. Therefore, it was banned and not allowed to be sold, so Wang Xu had to do it himself.

Moreover, the accumulation of this kind of technology is meaningful, and the significance is very large, so large that it can continue until decades later.

This includes operating systems, time-sharing systems, interactive systems, network routing, information exchange, and many more. If it develops smoothly, it will be the cloud of the future, and there is no need to change its path.

Wang Xu planned to use his own initial funds, and then take a look at the big sheep Shuimu to come up with a super simplified basic audio special version of this future cloud.

When Wang Xu explained the detailed requirements to several roommates, he saw disdainful looks from them all.

"What kind of looks do you have?" Wang Xu said angrily.

"You call this thing a supercomputer?" Boss Zhan said angrily, obviously feeling sorry for the disgusting lunch he just made.

"Super audio computer, or supercomputer for short, is there any problem?" Wang Xu frowned.

"Get out of here." Several roommates raised their feet to kick him at the same time.

The few of them had a fight. It was like this with college roommates. Everyone thought he was the father. Well, at least Wang Er thought so. They probably thought so too. This tradition seems to have not changed in future generations.

"Okay, let's stop making trouble. Just tell us whether you want to do it or not."

"It can be done, but we probably can't do it. There is too much content in your plan, and there is even an operating system. It's life-threatening. This is something that only a few students can do." This was Dadong's complaint.

"Yeah, do you really think we can do this with just a few of us?" Boss Zhan was a little unconfident. This guy usually pulls the boss like his second son, but he still gets timid when something happens.

"Is there anything I can't do? If I didn't have time, I could handle this operating system by myself, and it would probably only take a few months." Wang Xu raised his chin with a smile.

This is really not a boast, there is source code in this area in his skill infusion, well, it is the equipment maintenance technology. Just knocking out so much source code will kill people. Moreover, the source code is only suitable for special devices and cannot be copied, so if you want to realize it, you have to rely on people to pile it up.

"Just brag. You have to come up with a plan. Just talking about it won't do. If the plan is feasible, you can go to Uncle Qiang and report it to the superiors. I think considering how favored you are in the department now, you should apply for one. The project should not be a big problem.

But with so many people involved, and possibly across departments, it would be best to involve the Youth League Committee and the Student Union, as their organizational and coordination skills can still be of great help. "Even though Dadong is young, his vision is not low.

Back then, he was Mizuki, one of the top three players at the provincial level and recommended as a walk-on. Then he served as the League Secretary for another year. If he hadn't been lazy, he would have been able to join the student union of the department by now.

Oh, by the way, Uncle Qiang is the current Communist Youth League Secretary, named Wu Guoqiang. Because he is trusted by everyone, everyone likes to call him Uncle Qiang, or Uncle Qiang for short.

"Well, okay, I'll make a plan right now and start working on it quickly. Recently, I've been busy just working on materials, so I can't let the people in my hometown idle." Wang Xu raised his head and sighed, but was laughed at again.

In the next few days, Wang Xu spent time on this supercomputer project. Although it is a bit exaggerated to call it a supercomputer, but others don’t know it, and Wang Xu doesn’t understand it. After the technology inside is accumulated, it really is What a super calculating model.

——

(The first chapter is four thousand words. I am too lazy to divide it into one and two chapters. Please read a longer one if you are tired.)

Tap the screen to use advanced tools Tip: You can use left and right keyboard keys to browse between chapters.

You'll Also Like