Yang Zhengkun chose the second path. He was a little unwilling to accept that Baidu had not achieved any results in the project. If Baidu had continued to invest, perhaps the project would have been successful in a short time.

But he also considered the risks.

Commercial companies are all profit-driven, and this project is likely to face the same embarrassing situation as Baidu, with too much investment and too little output. The result of no one using it is that the project will eventually be withdrawn.

However, Yang Zhengkun did not change his mind. Instead, he firmly formed a team to overcome this problem and build a distributed database.

The so-called database is to effectively manage and access large amounts of data resources, improve data sharing, and enable multiple users to access data in the database at the same time; reduce data redundancy to improve data consistency and integrity, provide independence between data and applications, and thus reduce the cost of application development and maintenance.

At that time, the databases in the hands of Americans were all centralized databases.

A centralized database simply means that the storage and processing of data are concentrated on one computer.

The advantage is that most functions such as modification, backup, query, and access control are easy to implement.

The size of the database and the computer it is located on do not need to worry about whether the database is in a central location. Small businesses can set up a centralized database on a personal computer, while large businesses can have the entire database controlled by a mainframe.

The disadvantage is that if the central site computer or database system fails to operate, all users cannot use the system until the system is restored. In addition, the communication overhead from the terminal to the central site is very expensive.

If a large number of users extract data at the same time in a period of time, it will cause the server to crash.

Its biggest disadvantage is that the data it can store is limited. Once this database is full, you have to buy another one.

There is no such thing as win-win cooperation in the Western concept, only crazy plundering, and they will not stop until they squeeze the last drop of blood out of you.

Will Westerners who have mastered monopoly technology still ask others to skin them?

Take the world's largest database company, Jiaguang Wen, for example.

IOE - IBM minicomputer + Oracle database + EMC storage device, this is the database device with reinforced standards, and it has been the IT standard for Chinese enterprises for the past thirty years.

The cost of this set of hardware equipment is in the hundreds of millions.

Once one set of equipment is full of storage, you have to buy another set. Companies like Alibaba and Baidu had hundreds or even thousands of such sets of equipment at their peak.

Another disadvantage is that if you want to temporarily add some servers to handle more traffic, you have to place an order several months in advance.

There is also an unfair clause here, which is that once your company's data traffic decreases and the equipment is idle, you still have to continue paying annual usage fees.

If you have any comments, I'm sorry! Take it to the toilet, do you want it?

In this way, the concept of distributed database was proposed by someone, and it was proposed very early, in the 1970s.

A distributed database does not store data on one computer, but on countless computers.

Each computer has local application and autonomous processing capabilities, and can also participate in network-wide processing at the same time, which greatly reduces the pressure on computer work.

Twenty years later, people called this the cloud.

Distributed databases are unmatched by centralized databases in terms of processing speed and storage capacity, which greatly reduces costs.

But it naturally has its drawbacks, and there is still a major problem that cannot be solved: these computers connected in series cannot achieve financial-level stability. If one of the machines goes on strike, it is easy to cause a chain reaction like a domino effect - eventually the entire network will collapse collectively.

Technically, it's not difficult to ensure that if one machine fails, the other machines can immediately recover and back up the data. The challenge lies in ensuring that the workload on each machine is always balanced. If the workload is not balanced, one machine with an excessive workload is likely to crash, which in turn can trigger a major crash.

This is also the reason why distributed databases have never been successful despite being proposed in the 1970s.

This is what Yang Zhengkun did.

He has been working at Ali since 2010. Regardless of whether he was successful or not, Ali's Double Eleven server has never crashed.

It was when Wan Feng thought of Ali that he remembered who this person was.

Although his memory was a little fuzzy, Wan Feng was certain that this guy was the one who made 'Aliyun' for Ali.

Later, almost all the clouds on various websites were created by Ali.

Wan Feng's heart raced at the thought. Could it be that someone was guiding him? He had just casually searched and found his destined lover!

No matter how much it costs, we must get this guy to Huaguang! Even if it means destroying the salary structure of the South Bay Group, it's worth it.

This is a mountain of gold!

Someone was so excited that night that he didn't go to bed until dawn.

The result was that his long-standing habit of getting up at five in the morning was changed that day, and he slept until half past seven.

If Han Guangjia hadn't called him, he wouldn't have slept for who knows how long.

Wang Xuan did not break his promise. The next day when he arrived at Peking University, he called Yang Zhengkun to his office and discussed recommending him to work on databases.

At first, Yang Zhengkun looked confused: "Someone in our country wants to build a database? Is this true?"

"Of course it's true. I'll arrange a meeting for you two. You can have a good chat with that person. Maybe there will be something even more surprising to you."

Yang Zhengkun thought about it and nodded in agreement. It was just a chat! He didn't necessarily have to go.

So, at noon on July 1st, Wan Feng and Yang Zhengkun met again at Quanjude.

Yang Zhengkun has thin hair, is thin, and seems a little shy.

He doesn't drink or smoke.

This is a bit difficult to deal with, and it's a bit awkward.

Fortunately, Yang Zhengkun asked the question in advance and was straightforward.

"Ten thousand…"

"Professor Yang, you can just call me Xiao Wan."

"Xiao Wan! I would like to ask what your company does?" Yang Zhengkun is not particular about the name. If you say Xiao Wan, then call him Xiao Wan.

"Professor Yang! Our company does a wide variety of things, from machining to vehicle production to mobile phone and computer semiconductor material processing."

"So many? This is a group operation. I wonder what your group's pillar industry is?"

Pillar industry? Wan Feng had never considered this question.

What exactly is Nanwan Group's pillar industry? Initially, it was undoubtedly motorcycles, but as things have evolved, it seems to have become mixed, and now everything seems to be a pillar industry.

But you can't tell Yang Zhengkun that they are all pillars, that would be perfunctory.

"Mobile phones and cars."

Wan Feng picked out two of the group’s more representative industries and talked about them.