The Red Era, the Opening University Speech

Chapter 182 Semiconductor Lasers
In the evening, Grandma Li cooked herself, borrowed some mushrooms from her neighbor, and made a dish of mushroom stewed with flying dragon.

When the aroma was served, Li Mu, who had tasted all the delicacies from later generations, drooled.

The weather is gradually turning from autumn to winter, and it is already a bit chilly at night. A sip of chicken soup will warm you directly to the soles of your feet. The rich taste will make you praise it endlessly.

While eating, Li Mu continued to listen to General Zhao and Li Jun talking about life in the army, while thinking about Jin Shu's invitation.

It’s not for delicacies like tiger bone wine, three earth immortals, venison, orchid bear paws, but simply for the desire to see the three provinces that are the heavy industrial base of Xia State.

But he definitely won't have time to go there before the supercomputer is successful.

It’s hard to say what will happen next, I can only decline politely and look for an opportunity later.

While thinking about the problem, Li Mu quietly put the chicken legs that Grandma Li had taken from his bowl into the bowls of Li Lan and Li Yu without leaving any trace.

……

The next day, Li Mu got up early, washed up, and went directly to Beijing Institute of Technology.

Although he had already graduated, Li Mu was still a celebrity in the school. The security guard was impressed by him and greeted him with a smile, then let him in directly.

I arrived at the girls’ dormitory building and waited for a while before seeing Wang Ya, who looked sleepy, coming down from upstairs.

"Stayed up late again?" Seeing her expression, Li Mu knew that she must have studied until very late last night.

Researching supercomputers is a daunting task for everyone, without exception.

Hearing his voice, Wang Ya patted her face to sober herself up a little, and said, "It's okay, it won't affect the exercise."

"Really? If you are too sleepy, you can go back and sleep for a while. There is no need to force yourself." Li Mu advised.

If you exercise with a tired body, the effect will undoubtedly be greatly reduced.

Wang Ya shook her head. Although she looked tired, her eyes were very firm. "No need. Perseverance is the most important thing in everything. Now that I have started, I will definitely stick to it to the end."

"Okay, but if you feel uncomfortable anywhere, you must stop." Li Mu had no choice but to say.

He had a headache. His original intention was to let Wang Ya exercise well, but now it seemed to have the opposite effect.

……

Came to the Semiconductor Research Institute.

First go to the Hardware group to check the development status of the hardware.

Seeing Li Mu coming in and discussing with Wang Shoujue, Yan Peilin smiled slightly on his worried face, walked forward and said:
"There is a problem. The memory we currently use has great difficulty in meeting the storage capacity and storage speed requirements of supercomputers."

"If we were to use the previous tape storage, there would be problems with read and write speed, reliability, and physical size."

"For supercomputers, such memory is obviously not suitable."

After he finished speaking, he suddenly let out a long sigh.

How long has the magnetic tape storage been around? It's obviously not realistic to ask Li Mu to come up with a new storage device.

Seeing Li Mu's silence, Yan Peilin said, "Forget it, let's use the tape storage first, the slower the better."

"No need. I've considered the memory problem before, but it's still a little imperfect. Plus, Professor Huang's research hasn't been completed yet, so I haven't mentioned it for the time being." Li Mu waved his hand.

He had already downloaded the information, but the CD storage involved the production of semiconductor lasers, and there were not many people who could take on this task at present.

Hearing this, Wang Shoujue immediately said, "Oh, what's the solution? Professor Huang is busy developing a distributed projection lithography machine, so can you let me do it first?"

Distributed projection lithography machines are the basis for the development of supercomputers, and Huang Xinhua must not be allowed to be distracted from working on memory at this moment.

"Director Wang, your research task is not easy either!" Li Mu hesitated a little. He believed that Wang Shoujue could do it well, but the other party's task was already heavy enough.

Wang Shoujue laughed and said:
"You really underestimate me. Before you came to the Semiconductor Research Institute, I lived in the lab every day for research. Just leave it to me without worry."

"As long as we can develop a supercomputer as soon as possible, a little hard work doesn't matter."

"Besides, I dare not say that it is hard work. Professor Yan, Professor Huang, Professor Zhuang, and many other researchers are under much more pressure than me."

"Okay, thank you for your help, Director Wang." Li Mu pondered for a long time, then nodded.

……

After confirming the research plan, Wang Shoujue quickly gathered people and went to the conference room to continue the meeting.

But not many people came.

Now everyone has research tasks on hand and is working around the clock. Less than ten people are free to participate, and eight of them are graduate students.

Wang Shoujue walked into the meeting room and was a little embarrassed to see so few people. He said to Li Mu on the stage: "Don't worry, I will continue to apply to see if I can transfer more people."

"It's okay, Director Wang, you go ahead and do it first." Li Mu smiled and didn't care.

He knew that Wang Shoujue must have transferred all the people who could be transferred.

Now, all graduate students at Kyoto universities whose majors involve semiconductors and computing technology are basically here.

If we want more people, we can probably only choose from college students.

He looked at the audience, greeted them politely, and then got straight to the point:

"Before you came here, you should have heard Director Wang talk about the problems we encountered, so I won't waste any more words."

"Let's talk directly about the memory we want to develop."

"My idea is to use an optical method to read and store data from optical storage media, which mainly relies on semiconductor lasers as light sources."

"The overall physical structure mainly includes the optical disc, optical reading device, mechanical device and electronic control part."

……

Li Mu spoke freely on the stage.

The professors and graduate students below were taking notes quickly, stopping from time to time to ask questions.

"Advisor Li, could you explain in more detail what semiconductor lasers are?" a graduate student raised his hand and asked.

Li Mu said:
"Of course. A semiconductor laser is actually a laser that uses semiconductor materials as its working material."

"The electrons are used to emit light when they transition between energy bands, and the cleavage planes of the semiconductor crystal are used to form two parallel reflective mirrors to form a resonant cavity."

"Then, the light is oscillated, fed back, and the radiation generated is amplified to output laser light."

"Its basic structure consists of three parts: active region, optical waveguide and resonant cavity. The active region is the core area for generating lasers, usually using a double heterojunction structure, which is used for..."

……

Through questions and answers, all the participants in the meeting established an understanding of semiconductor lasers and optical disk storage in their minds.

According to today's content, they will not encounter too many problems, at least in the short term.

After Li Mu finished speaking, Wang Shoujue immediately stepped forward and said with a smile: "If what you said is imperfect, then what can be called perfect?"

"It doesn't mean that we have to write a plan that doesn't even require us to think about it to be perfect, right?"

These words made the professors and graduate students nearby couldn't help laughing.

They had come with the idea of ​​completing the task at all costs, but after a meeting, they felt that the memory problem was nothing special. What was the difficulty? Just follow the instructions.

Of course, everyone knew that this was not simply a memory issue, but Li Mu had opened up a smooth road for them.

Seeing everyone's confident look, Li Mu did not open his mouth to criticize them.

The problem of optical disk storage is not so easy to solve. It may not be difficult to make it, but the difficult part is how to solve the problems of not being able to work normally at room temperature, poor heat dissipation, and low photoelectric conversion efficiency.

……

Soon after the meeting, the development of optical disk storage began rapidly.

Although there are still some shortages in manpower, there are no major problems in dealing with current research.

By the time the development of optical disk storage enters the critical stage, Huang Xinhua's distributed projection lithography machine should also be completed, and manpower will not be a problem.

……

Li Mu was not idle either. After separating from Wang Shoujue, he continued to go to other research and development teams to solve problems.

First come to the software group.

Software is actually more important than hardware.

Some people may think that it is inappropriate to consider software when there is no computer yet.

But without software, no matter how good the hardware is, it is just decoration.

It's like a Gundam. It's powerful, but without a cockpit, you can't move it no matter how good its performance is.

In the later development of computers in the Xia Dynasty, software has always been a major flaw. The country has computing speed but no independently developed application software.

In this case, if foreign countries restrict the export of high-end and key application software to Xia, the losses will undoubtedly be very heavy.

Therefore, in Li Mu's plan, software research is no less important than hardware.

Software development is not easy.

Although China has the J language and is leading the world in computer programming languages, there is a problem that cannot be solved in a short period of time.

That is talent cultivation.

From the birth of J language, to the emergence of software concepts, to the research and development of supercomputers, almost all of them were led by Li Mu.

Even though Huang Kun and others have received some training in the process, it is still very difficult for them to solve the software problem on their own.

……

Software group.

As the deputy team leader, Huang Kun was under a lot of pressure. In just a few days, he lost a lot of weight visibly.

But even though he was very tired, he still insisted on staying at the forefront of research, leading the team to solve the current problems one by one according to the direction given by Li Mu at the meeting yesterday.

When Li Mu arrived, Huang Kun had just finished solving a problem on a parallel algorithm. He rubbed his swollen temples and decided to continue working.

"Take a break." Li Mu walked forward, picked up the water cup on the table, handed it to him, and took the problem report in his hand.

After taking a look, he said, "To complete the design of job scheduling software, we must first consider how to reasonably allocate limited computing resources among multiple tasks."

"Your idea of ​​dividing it into four modules, task insertion, task selection, task scheduling and task processing, is a good one."

"However, it should be noted that the task scheduling algorithm needs to dynamically adjust the task allocation strategy according to the current load of the system, as well as handle the expiration of tasks. For some tasks with high real-time requirements, the level can be increased to a limited extent."

……

After listening to Li Mu, Huang Kun's eyes revealed a touch of admiration and emotion, and said: "You are worthy of it. You can solve this problem so quickly."

"The job scheduling system development team has been stuck on this issue for a long time."

"In the final analysis, it's because we haven't trained enough talents." Li Mu shook his head and said, "You or the people in the J language application research group can solve this problem by spending some time."

"I believe that the reason why you have been kept here until now is not because you are lazy, but because you are solving other more important problems."

Hearing his words, Huang Kun smiled bitterly.

When he was the deputy team leader of the software group, he was eager to do a big job, but he didn't expect so many problems to arise as soon as he got started.

Although the study of J language has gradually become popular in universities in China, most of them are theoretical and lack practice.

Now, the only people who can complete the research independently are a few people from the original J language application research group and some knowledgeable professional scholars.

Others can only do some less important tasks for the time being, and even so, small problems often arise.

You can imagine how difficult it is to develop software with them.

But even so, facing Li Mu's concern, Huang Kun did not complain at all, and said: "It's okay, we can handle it."

"I know you can do it." Li Mu smiled helplessly, "But the research task cannot wait for you to slowly solve it."

After thinking for a while, he continued:

"In this way, I will propose to Mr. Yan later to let him recruit some college students to supplement the manpower."

"We can put the research on hold for now. Let's organize a short-term training course first."

"I will also teach. Let's train people first, at least junior talents who can solve simple problems, and then continue to tackle tough problems. Otherwise, you will fail before the research is completed."

Huang Kun pondered for a long time about Li Mu's proposal and nodded helplessly.

Now was not the time to act on impulse, reason told him that this was the best choice.

……

When we left the software group, it was already completely dark.

I first went to see Yan Peilin and told him about the training class. The latter nodded and agreed immediately without any hesitation.

However, it will take at least a month to complete the selection and training of personnel, and the software team's task in the short term is still very arduous.

……

Walking out of Yan Peilin's office, Li Mu looked at the time and it was already 9 o'clock in the evening.

"Let's stop here for today." He thought for a moment, went to the institute's canteen to eat, then rode his bicycle back to his dormitory at USTC.

In fact, it would be more convenient to stay in the Semiconductor Research Institute, but the huge supercomputing research team had already crowded the institute, and Wang Shoujue had six people living in a dormitory of less than 10 square meters.

He had taken a look before and felt that there was no need to force his way in.

……

Go back to dorms.

When General Zhao saw him come back, he greeted him and handed him the hot water.

"Thank you, Brother Zhao." Li Mu took the hot water, washed up simply, and immediately sat down at the table.

(End of this chapter)