Understand the relationship between Filecoin and IPFS, as well as the risks you need to pay attention to when participating in mining

Original title: "FL in-depth analysis | Is Filecoin really the next gold mine?"
Author: Fan Xuxin

What is Filecoin? How is it related to IPFS?

IPFS is a network protocol with a series of excellent features such as content addressing and version control. It ensures the safe storage and transmission of files in P2P networks. IPFS is considered to be the next generation of network transmission protocol and will have the opportunity to replace the HTTP protocol. Filecoin is a decentralized storage system based on the IPFS network protocol. Both IPFS and Filecoin are developed by Protocol Labs. Filecoin adds a market mechanism to regulate storage costs on the basis of the IPFS protocol, and also adds an incentive mechanism to ensure the stable operation of the system. In the Filecoin system, users can choose services with corresponding prices based on their own conditions such as speed, number of copies, and budget.

Filecoin uses the EC (Expected Consensus) consensus, which is similar to the PoS consensus, but the difference is that Filecoin determines the probability of packaging rights based on effective storage rather than the number of pledged tokens. Effective storage is one of the main features that distinguishes Filecoin from other similar projects. In order to determine effective storage, Filecoin introduced PoRep (Proof of Replication) and PoSt (Proof of Spacetime) proofs, where PoRep is used to prove that a node stores a copy of the corresponding file on the physical storage, and PoSt is used to prove that a node stores the corresponding file within a certain period of time.

What is the prospect of Filecoin?

To talk about the prospects of a new product, it is nothing more than considering the two perspectives of cost and experience. Compared with its competitors, does it have lower costs or better user experience? There is no doubt that IPFS is the best decentralized storage protocol at present. Compared with the existing centralized file system, it has a series of advantages such as content addressing, version control, and non-tamperability (this article focuses on the relevant content of Filecoin mining, and the specific advantages are not discussed in detail. If you are interested, you can learn more about it through the participation materials at the end of the article). The IPFS protocol can ensure the safe and efficient storage and transmission of files in the P2P network, but to be applied on a large scale in the P2P network, the system needs to be able to self-motivate and operate. At this time, it is necessary to introduce the corresponding mechanism to achieve this, and Filecoin came into being.

In the traditional blockchain field, the cost of information storage is extremely high. For example, the cost of storing 1MB of data on Ethereum can reach tens of thousands of US dollars at its highest. Therefore, this has greatly hindered the development and application of DApps. Among the projects presented at the 19th Berlin Web3 Conference, storage projects accounted for more than 10%, which is enough to illustrate the urgency of solving storage problems in the blockchain field. Among all decentralized storage projects, Filecoin is undoubtedly the most popular, and IPFS has been running stably for two years, which fully proves the robustness of its underlying protocol, but the actual effect of Filecoin still needs to be tested in practice. Of course, there are also some blockchain storage projects that are improvements made to centralized cloud storage services, such as Siacoin, whose storage cost is only 1/10 of centralized storage, but as of now, the data stored on the Sia platform is only 2PB. The main reason is that the user threshold is too high and the experience is not good. Whether Filecoin can occupy the market share of traditional centralized storage depends on its ultimate user experience and its storage price.

At present, most of Filecoin's nodes are deployed in IDC computer rooms. The actual deployment and operation costs are not lower than those of traditional centralized cloud storage, and the storage process is more complicated (PoRep and PoSt need to be run, which involve a large number of zero-knowledge proof operations), resulting in higher actual storage costs. Therefore, in the early stages of Filecoin's launch, the actual data storage cost may be higher than centralized cloud storage, but since the initial nodes lower the storage price in order to obtain block rewards, the actual storage quotation may be lower than that of traditional centralized cloud storage. In the long run, Filecoin still needs to further leverage its P2P storage advantages, transform storage devices from professional to civilian, and optimize its algorithms to reduce storage costs without affecting user experience.

Storage is an important issue that needs to be solved in the blockchain field, so a large number of storage projects were demonstrated at the 2019 Web3 Summit. IPFS is an important part of the Web3 vision, and its development will affect the development of Web3 to a certain extent. Similarly, the development of Web3 will also determine the future of IPFS to a certain extent. Filecoin is a storage project based on the IPFS protocol initiated by IPFS officials, and it is undoubtedly highly anticipated.

What is the Filecoin mining process?

As shown in the figure below, Filecoin is mainly composed of three parts, namely the storage market (on-chain), the Filecoin blockchain, and the retrieval market (off-chain). The storage market and the retrieval market are on-chain and off-chain respectively for security and efficiency considerations. For users, the storage frequency is relatively low and the security requirements are relatively high, so the storage process is carried out on the chain. When there is a certain amount of data, the retrieval frequency is much higher than the storage frequency. Considering the performance of processing data on the chain, the retrieval process is carried out off-chain. In order to solve the payment security problem during the retrieval process, Filecoin adopts a micropayment strategy. Simply put, the process is to divide a file into many copies, and the user pays the corresponding fee for each copy of the data.

The types of miners corresponding to the two major markets of Filecoin are storage miners and retrieval miners. Storage miners are mainly responsible for data storage and block packaging, while retrieval miners are mainly responsible for data query. In the future, after the Filecoin mainnet runs stably, repair miners will be introduced to be mainly responsible for data maintenance.

Different from what is shown in the above figure, in the early version of Filecoin, the order matching mechanism was not implemented in the storage market and the retrieval market, but the order grabbing mechanism was adopted. There are three processes corresponding to the three major parts of Filecoin, namely the storage process, the retrieval process, and the block packaging and reward process. The simplified process and miner income are shown in the following figure:

From the above, we can see that unlike traditional PoW mining, the Filecoin mining process is much more complicated, and an important factor that determines the early mining revenue is effective storage. Effective storage is a core feature that distinguishes Filecoin from other decentralized storage projects.

In Filecoin's EC consensus, effective storage is similar to the rights in PoS, which determines the probability of miners obtaining packaging rights, that is, the proportion of miners' effective storage in the entire network storage is proportional to their final mining income in probability. The factors that determine effective storage are shown in the figure below, including storage quotes, network conditions, CPU performance and other hardware performance. Of course, higher effective storage can be obtained under the same hardware conditions by optimizing the mining algorithm. However, it is still unknown how much revenue can be increased by algorithm optimization.

In the EC consensus, effective storage is similar to the rights and interests in the PoS consensus, and in Filecoin, it is not the effective storage of a node in a single block cycle that determines the probability of packaging, but all the effective storage since the mainnet went online. Filecoin will have a certain Matthew effect, and the earlier you participate, the more advantages you have. However, unlike the PoS consensus, storage on Filecoin is a non-reusable resource. Therefore, when a node's storage space is full, the effective storage will no longer grow. To obtain more benefits, it is necessary to further increase the corresponding hardware, and at this time it becomes a "latecomer."

Filecoin token release

Unlike most projects, Filecoin's token release strategy is linear release. The tokens of the team, foundation, and ICO are all released linearly, that is, each part of the tokens is gradually released with mining in each block. The amount of Filecoin mining release is also different from other projects. Filecoin mining rewards are halved every 6 years, and the halving amount is calculated by the following formula:

After calculation, Filecoin's initial daily block reward was 440,000, and after one year, it dropped to 395,000, and at the end of the sixth year, it dropped to 220,000, completing a linear halving. This halving method is undoubtedly very beneficial to miners. Based on the above information, the token situation of Filecoin in the entire network within 6 years is shown in the figure below (since the linear release and unlocking period of private placement and ICO varies from 6 months to 3 years, the average value is taken here to be released in 1.5 years):

As can be seen from the above figure, due to the adoption of a linear release strategy, Filecoin tokens were very scarce in the early stage, and because participation in Filecoin requires a certain number of Fil tokens to be pledged, further scarcity is caused. Therefore, it can be predicted that the price of Fil tokens will be relatively high in the early stage, but as the tokens are continuously released, the price will continue to decline. The actual token price trend may not be completely consistent with the above prediction, because in addition to the number of tokens, market heat and supply and demand relationships will also affect the price. However, it should be emphasized that in the long run, the number of tokens is one of the most important price influencing factors.

Filecoin payback period and revenue calculation

First of all, it should be emphasized that Filecoin has not yet been launched, and it is relatively difficult to calculate the payback period, and the accuracy cannot be guaranteed. However, it is still possible to make a rough estimate of the income. As shown in the following table, the following table shows the approximate release of tokens in the first year of the main launch (since the initial mining release in the first year is 440,000, and the mining release at the end of the first year is 395,000, so the average value is 420,000).

It can be seen that in the first year of the mainnet launch, the number of Fil tokens released every day is about 1 million, and 30 million every month. Assuming that the number of tokens produced by a single T computing power in the entire network is the same, the payback period can be calculated by the following formula.

Among them, the price of computing power is the quotation of the mining machine manufacturer. The daily mining output of the entire network is determined to be about 420,000 coins. You only need to substitute the expected coin price and the expected total network computing power into the formula to estimate the payback period. Regarding the coin price, this article does not make too much analysis. The official estimate of the total network computing power by Protocol Labs is 100EB, but whether it can actually be achieved remains to be seen. As for the income, it can also be roughly estimated in the same way.

What issues should I pay attention to when participating in Filecoin?

1. The transparency of mining machines or computing power

Currently, there are two main ways to sell Filecoin mining machines on the market, namely direct sales of mining machines and sales based on computing power. The transparency of mining machine sales is relatively high, and the corresponding prices of similar hardware can be found on e-commerce platforms.

The sale of computing power refers to the way of quoting and selling at a price per T. This is to some extent inspired by the pricing method of PoW mining, but there are very big differences between the two. Taking Bitcoin as an example, Bitcoin computing power is priced in T, and its unit is H, which is the number of times the Hash algorithm is run per second. This unit is directly related to the mining chip, the core component of the mining machine, and the Bitcoin main network has been online for many years, and the transparency of the mining machine is already very high. Filecoin is priced in T, and its unit is B, which is commonly referred to as bytes. What is directly related to this unit is the hard disk space of the mining machine. It has been mentioned many times above that the key parameter that determines Filecoin mining is effective storage, which has little correlation with the storage space of the mining machine itself. The factors that determine whether the storage space can be converted into effective storage include a series of software and hardware conditions such as CPU performance and network conditions. It is still unknown how much output each manufacturer can produce with a single T computing power.

Of course, it does not mean that the method of pricing by T is completely incorrect. Under the premise of high transparency (being able to know how much hardware is behind each T of computing power), the method of pricing by T is also reasonable. However, before the launch of the Filecoin mainnet, due to the lack of transparency of information, there are many ways for manufacturers to manipulate the pricing by T, so it is necessary to be very cautious when purchasing computing power by T, and try to ensure the transparency of information before participating. After the launch of the Filecoin mainnet, no matter how much hardware is behind a single T, the price can be evaluated directly by calculating the output of a single T. At this time, pricing by T will be very intuitive.

2. Testnet mining issues

After the Filecoin testnet is launched, whether there will be rewards for participating in mining, or how much the rewards will be, the specific plan has not yet been announced. At present, Protocol Labs has just announced the mainnet launch plan, with the testnet to be launched on December 11 and the mainnet to be launched in March 2020. However, from experience, it is unlikely that mining will be carried out directly at normal proportions on the testnet. For example, mining on Kusama, the testnet of Polkadot, has no direct relationship with the Polkadot mainnet. For example, mining on the testnet of Nervos only takes a certain amount of mainnet coins as mining rewards. Therefore, after the Filecoin testnet is launched, it is not recommended to rush to participate in mining.

3. Impact of algorithm optimization on mining efficiency

Frankly speaking, this question is currently very uncertain. Since Filecoin is different from PoW mining and the algorithm accounts for a high proportion, algorithm optimization will have a certain impact on mining efficiency, but how big the specific impact will be and whether it is consistent with the claims of various manufacturers will remain unknown before the mainnet is launched. This will also be the biggest uncertainty factor for early investment. What is certain is that due to the survival of the fittest in the market, the computing power output level of the entire network will eventually tend to be basically consistent, so the advantages of algorithm optimization are mainly concentrated in the early stage, and as the market matures, its advantages will gradually be weakened.

Summarize

The storage problem of blockchain needs to be solved urgently, and Filecoin is the project with the highest expectation to solve this problem. In addition, the IPFS protocol has been running stably for more than two years, so Filecoin itself is very worth looking forward to. However, before the mainnet is launched, the information is very opaque and there are many traps. Therefore, if you want to participate in early mining, please be sure to conduct sufficient investigation and research.

Although there is a certain Matthew effect in Filecoin mining, and mining the "first mine" has certain advantages, before the main network is launched, the risk of participating in mining is very high, because the information uncertainty and opacity are very high at this time. Therefore, whether Filecoin is the next gold mine still needs time to test, but what is certain is that if you participate in mining without sufficient information, the risk is far greater than the benefit.

Source link: mp.weixin.qq.com