Video|"8" BTC.com CEO solemn: the trouble of the number one player

Solemn, BTC.com CEO, industry senior technical expert. During his Ph.D. degree at the University of Science and Technology of China, he went to Yale University in the United States as a visiting scholar. Zhuang Zhong joined BitChina in 2015 and is currently leading BTC.com to explore more blockchain related products and services.

Cover of the 22nd issue 2

In 2011, Zhuang Zhong heard about Bitcoin, and the registered account transaction was in 2012. In 2013, Pumpkin Zhang developed the first FPGA mining machine, which opened a new era of FPGA mining.

"At that time, the scale and professionalism of the FPGA mining machine could not be compared with the current ASIC mining machine. The ASIC mining machine is undoubtedly an epoch-making product, and its computing power and efficiency are greatly improved compared with FPGA. But in 2012 The emergence of FPGAs makes people feel that there are still many people in the world who are paying attention to this matter."

Zhuang Zhong’s previous dream may be to enter colleges and universities for academic work, and his first formal job is in the digital currency industry – joining the roast cat mining machine, responsible for self-propelled mining, and some monitoring and management of the mine.

“At that time, the whole industry was still very early, and the choice was small. There were only a few BTC China exchanges. Secondly, individuals were interested in mining. It was natural to learn about Bitcoin after they first learned about Bitcoin. Being attracted by mining," said Zhuang.

At the end of 2012, the entire Bitcoin network was less than 10t, and now a mining machine is more than the entire network. Bitcoin security must rely on high network-wide computing power to ensure that the system is not easily controlled by a small group of people. The best way to do this is to produce more specialized equipment and create a very high computing power. Through the market-oriented direction of these devices, the Bitcoin network cannot be easily controlled by some institutions and individuals. This is also a simple reason why Zhuangzhong is attracted by the mining machine industry.

So, what is the mining that makes Zhuangzhong fascinated?

"In principle, mining is to let the miners get the billing rights in the bitcoin system through the workload proof mechanism, and they can get the billing right to speak in the system, so they get rewards. But it takes a cost to speak in the system. You can limit the node to do evil. The most common is the distributed system, you can fake a lot of nodes in the network. But in the bitcoin system, you want to make changes to the state in the system, you have to pay a lot. Out of computing resources, so even if you have a lot of fake nodes, you can't do enough power, this is the role of the workload proof mechanism." Solemn answer.

WechatIMG512

Mine pool business is to watch the day to eat

At present, some of the mine's code on the market is open source, and its business model can be replicated. Therefore, there is no more innovation in the business model of the mining pool. The main competition between the current mining pools is in technology, computing power and capital. Among them, the most important is the calculation power. The more the calculation power, the larger the block rate, the more stable the cash flow of the mining pool. The competition on computing power seems to be the biggest competition point between the mining pools.

"There are actually two points here. First, because most of the mines are now in the PPS mode, the PPS model means that the mines have to bear the fluctuations caused by luck. No matter how many blocks the mine has dug today, we have to follow the users. The corresponding calculation power gives him the proceeds. Once the mine pool is not lucky, it is a state of loss. All the mining pools will face the problem of reserve funds regardless of the size of the calculation. You need to have a certain reserve of funds to cope with this. The change in cash flow caused by the lucky value. This value itself has nothing to do with the ratio of the calculation power, but the overall stability of the mining pool with high computing power will be relatively high, and its lucky value will not appear relatively large. The deviation, the cash flow is more stable. If the mining capacity is relatively small, it will be long-term loss. Of course, in the dimension of the year, even if a mining pool only has 23% of computing power, his The probability of a lucky value for the year is also above 97%." Zhuang said.

However, there is still a 10% probability that it will be lower than 97%, which is dependent on the power ratio. However, for the reserve fund, the mining pool is computationally intensive, and the greater the jitter of the lucky value every day, the greater the loss to the mining pool. Therefore, regardless of the size of the mine pool, it is necessary to prepare almost the same reserves in the reserve fund. From this point of view, the mining pool business really has a feeling of watching the world.

From the perspective of income, there will be very large fluctuations every month. But from the year of view, most of the mining pools are still relatively stable. Of course, this is more for the currency like Bitcoin, which has a long block and block time interval.

For currencies with relatively short block intervals, the frequency of the mining pool with higher computing power is higher than the average yield rate of other mining pools, and the lone block rate is relatively lower. This effect is relatively weak on Bitcoin, but In some currencies with short intervals, this value has largely affected the performance of the ore pool. Therefore, for the currency with faster output, the greater the proportion of computing power, the lower the lone block rate, and the more advantageous it is for the mining pool. Correspondingly, it means that the mine pool can give customers a lower rate and can form a virtuous circle.

WechatIMG517

Block attack is unpreventable

On June 12th, the big pool of the scorpion fish in Weibo issued a document accusing a mine pool of a quarter of the power to attack other mines. Such an attack would cause some small mines to lose money, even collapse. We call this kind of power-attacking behavior that harms others unfavorable as a "block attack", and the industry seems to be helpless at this behavior.

"The Tibetan block attack has no way to completely prevent it in principle. This is a game of attack and defense. Those who do Tibetan block attacks can attack in many ways, and they can hardly find out. He reported the block. But he did not submit the task of the block to the mine. For the attacker, he only lost a very small amount of money. But for the mine, it directly lost the income of a block. Behavior, the attacker himself can not get extra benefits, purely an attack behavior." Zhuang said.

Technically, we can do some preventive measures. The most basic is to monitor the user trust value. Once a user with a long time trust value is suspected, it is necessary to consider that the attacker may adopt Attack behavior and manner.

Do some monitoring of the user's power behavior, which can only be a defense. If the attacker keeps changing IP addresses and splits the power into very small or very random calculations, it is more difficult for the mine to find such users.

Another option is at the mining level, we are working with miners to develop some safety components to prevent this problem from the mining level. But now many mining machines are also facing the problem of this kind of virus attack, especially the early version of the mining machine. Because early users did not know that the software behind the operation of the mining machine would cause a block attack. If you can do security reinforcement at the mining machine manufacturer level, and ensure that the mining machine is submitting the task, to do this, in fact, it has been a good way to alleviate the problem.

Of course, everyone also thought about hard-forking bitcoin, and completely avoiding this problem by hard fork, which is also the best solution at present. But this difficulty is imaginable. Now soft forks are a difficult thing to push, let alone hard forks?

Relatively speaking, there is currently no such large-scale organization on Bitcoin. But if such a thing is really done, it will really make all the mines a headache. If the block attack really becomes a pain point for the entire industry, most of the mines may be forced to return to the PPLNS model.

WechatIMG514

Is there really a 51% computational attack?

There is a saying called "Bit Continental Mine Pool" and "Other Mine Pool". This means that there are only two types of mining pools in the mining industry. One is attributed to the Bit Continental, called the “Bit Continental Mine Pool”, and the other is not attributed to the Bit Continental, called “Other Mine Pool”. This is actually the right Description of the trend of centralized power calculation in mining pools.

"I think the definition of [Bit Continental] is quite strange in itself. I know that everyone might want to say that BTC.com, ANTpool, ViaBTC, BTC.top are all bite continental mines, but bit mainland can't control at all. The latter two, even difficult to express their opinions in their decision-making." Zhuang said.

The pool itself does not control the power, but there is a systemic risk, which means that these pools are likely to add up to 50% in a short period of time. If you get together, what you might do in the short term.

But once this problem is discovered, someone will migrate to other mines. This means that a mine pool share exceeds a certain level, which will lead to certain emotions in the market. We do not want to have such systemic risks. We welcome the true drive optimization agreement, which means that the mine allows users to choose their own. The transaction in the memory pool is packaged.

But because of efficiency issues, this agreement is not currently widely adopted.

Now someone is designing a new agreement, hoping that users can freely decide the deal of the pool packaging, which will largely prevent the mine from doing evil. If implemented, many mine pools will welcome and embrace this change. At present, the upgrade of the Bitcoin network protocol is becoming more and more independent of the miner's voting mechanism, which means that the user can freely choose the mine that he believes is the best and most stable. Of course, there may still be a 50% computational attack problem, but this situation has already improved a lot.

WechatIMG513

Bright surface, behind the bitterness

When asked about the possibility of a broken mine, the solemnity is that the break may be small.

Bitcoin's mining machines are undoubtedly more competitive in the market. It is not difficult to find that the leading mining pools on the market, except for SlushPool, have very strong IP mine users in the early days, purely tap water, more mine pools. Each has its own channel to get customers.

Many of them are bound to their own miners, they have a good relationship, and there are many mines that have a good relationship with the transaction. In fact, many mines have more or less affiliated businesses in their business. . Although everyone seems to think that the mining pool seems to be more glamorous, in fact, more behind it is very bitter.

First of all, the mines that are completely independent in the true sense are actually very few. There are other businesses behind the big mining pools that are currently active on the market. When the user chooses the mining pool, the first is to consider the rate, and the second is the stability of the service quality.

In addition, the mining pool has to fight for the ability to help users integrate resources. That is how to help the miners solve the problem of buying a mining machine and converting the digital currency into income. There are many points in the whole chain that need the mine pool to integrate resources upwards and downwards and provide them to users.

In addition to the mining pool, BTC.com other businesses. Such as browsers, data services, etc.

The browser business has been exploring. The biggest difficulty of this business is that the blockchain browser is a partial tool application. It is difficult to benefit from this business and maintain the healthy development of the business.

In terms of data services, more professional organizations, or related service providers, are found to be interested in this piece, but the market as a whole is relatively small.

In fact, the data of the entire industry is still in a stage to be explored, because the value of data is often obtained by the interaction between the user and the provider of the data. We may not know what the value of this data is, and the user does not necessarily know that we can provide such data, so there is still a big obstacle here. The other is that it has high operating costs, and how to balance this business development is a big challenge.

Our core is to optimize the service of blockchain data, so that it can better support more currencies. We also want to reach more professional users and provide them with more flexible data platform services. We will discuss this in this section.

Our long-term impression is that we focus on the Bitcoin mining pool, but our positioning is to become the entrance to the blockchain industry. At present, the mine pool and blockchain browser are everyone's attention, but our business is not limited to this.

In the browser business, the introduction of new content business, enriching the products on the browser, we are also brewing a larger product, ready to present to everyone this year.

The mining pool is optimizing the product. Recently, a Bata version of the mineral pool product interface was launched, and users are expected to go to the test. In addition, we will work with third-party financial businesses to integrate some new features into our products to provide users with a more comprehensive financial-related business. This is expected to be launched in the second half of this year.

Text|Jia Xiaobe video|"8 questions" column

Article tail map