The launch of Ethereum 2.0 will be a major milestone event in the encryption community, which will determine the development trend of the blockchain industry to a certain extent, and will also create many opportunities for participation, which will have a huge impact on the entire encryption ecosystem.

This article will try to understand some basic frameworks of Ethereum 2.0 from the perspective of participants, and look forward to the ensuing market opportunities.

Summary of Ethereum's development stages

The development stage of Ethereum is divided into four parts, namely:

• Frontier - launched in July 2015;

• Homestead - launched March 2016;

• Metropolis - the first phase of Byzantium will be launched in October 2017, and the second phase of Constantinople will be launched in February 2019;

• Serenity Serenity – Expected to be live by the end of 2020.

When the first stage of Ethereum’s main network was launched, all aspects of functions were not complete, so each stage of development tried to solve certain problems, such as usability, performance improvement, and graphical interface optimization. The first three stages all use PoW-type consensus, and in the serenity stage, it will be converted to PoS-type consensus. The Serenity Serenity phase has been referred to in the industry as Ethereum 2.0 for the past two years and represents a very significant upgrade.

The biggest difference between the upgrade of Ethereum 2.0 and the previous upgrades is that this upgrade is not a hard fork in the traditional sense, but starts with the migration of Ether from the 1.0 chain to the 2.0 chain. User migration is voluntary, so there will be two chains at the same time, one is the Beacon chain of PoS, and the other is the current ETH1.0 chain. According to the roadmap, the two chains will continue to be developed separately, and then they will eventually merge into one. This is a clever and realistic method (it is also a helpless move considering the sustainability of the huge ecology of ETH1.0).

The main purpose of the Beacon chain is to complete the realization of PoS and sharding, and to solve the expansion problem. The 1.0 chain will continue to evolve to 1.x, to complete the realization of stateless Ethereum, and to solve the problem of state explosion. Ultimately, it is for the scalability of Ethereum. As Vitalik said, ETH2.0 is all about scaling.

The three phases of ETH2.0 described by Consensys:

Ethereum 1.0 to 1.X (Development of Stateless Ethereum)

The destination of Ethereum 1.0 - the 1.x state becomes the 2.0 shard chain

One of the remaining purposes of ETH1.0 continues to exist as the basic chain of the current Ethereum ecosystem. The primary consideration in the 2.0 upgrade process is that there are still a large number of Dapps on the 1.0 chain, and a large number of developers are conducting various product/optimization/underlying/expansion research and development around 1.0.

However, the 2.0 Beacon chain has no way to transfer Ether at the beginning. The transfer can only happen on the shard chain, and the sharding can only be realized in Phase 1 of 2.0, and the smart contract will only be restarted in Phase 2, so these smart contracts It can only be migrated in Phase2. Therefore, 1.0 also ensures that the current ecological development of Ethereum will not be affected.

On the other hand, 1.0 will continue to evolve to 1.x, so that the two chains of Ethereum will start to merge after completing their respective tasks. Therefore, the complete ETH2.0 includes not only the three phases of the quiet stage, but also the current research and development results of ETH1.x, until they are combined into one.

The current plan is that ETH1.x is merged as a shard chain of 2.0, and the state is completely migrated to 2.0, so that the current ecology can be directly migrated without being affected.

Ethereum 1.x shoulders the research and development of stateless Ethereum

Ethereum uses the MPT (Merkle Patricia Trie) data structure, but with the increase of blocks, the amount of stored data is also increasing, which will inevitably cause a state explosion in the future.

Every new transaction needs to traverse the MPT tree, which is a heavy burden for scalability. Therefore, the concept of statelessness has been proposed, that is, when the client node does not need to save the state, it can still verify the transaction. The previous state can be realized through the state witness.

So in 1.x, there will be two types of important roles, one is the block proposer (block proposer), which contains transaction information, and all the status and witness information that previous transactions need to access. The other type is the state witness (state provider), which saves all previous state information to provide transaction verification.

New roles require new incentives. Ethereum Foundation researchers Sam Wilson and Ansgar Dietrichs discussed three incentive models for state witnesses. It is still relatively early, and the discussion of the three types of models is listed here:

In addition to stateless Ethereum, developers have also discussed the mode of state rent to limit the growth of too large a state, but developers have also discussed the mode of state rent, but it may have a greater impact on existing smart contracts and Dapps (no Paying rent will lead to the deactivation of the contract), which requires a hard fork to be realized, which will cause great controversy, so the current research on state rent has been put on hold, and resource-oriented development is still stateless or semi-stateless.

In short, the research on stateless Ethereum is not only a 1.x issue, but will eventually become the general model of Ethereum 2.0. The Ethereum Foundation believes that the shard chain of Ethereum 2.0 must be stateless in the future .

Problems to be solved by Ethereum 2.0

Ethereum 2.0 has four issues that need to be addressed: fork selection, finality, sharding, and validator responsibility. And use different technical and institutional means to achieve it.

Fork Choice - Guaranteed Security

Ethereum 2.0 uses LMD GHOST ("Latest Message Driven Greedy Heaviest-Observed Sub-Tree") as the fork option. This includes two protocols: LMD and Ghost:

Ghost is a very mature protocol on the PoW chain. It is a principle of the longest chain. Compared with the longest chain principle on BTC, Ghost chooses the protocol with the most subtrees, so the longest chain may not be Ghost The selected chain. Compared with the longest chain principle, the Ghost principle can achieve: 1. Convergence; 2. Anti-51% attack; 3. TPS is similar to the longest chain, but the security has not decreased. The Ghost protocol is also currently used by Ethereum 1.0.

LMD is a newly added fork principle in 2.0, and together with Ghost, it forms LMD GHOST. LMD means latest news. Since PoS was used in 2.0 to join the role of the verifier, the latest news refers to the proof of more verifiers, that is, in the Ghost principle, adding more verifiers' attestation (attestation), which makes the longest chain Becomes a canonical chain with finality according to the Casper protocol.

Finality problem - will not be reversed, solving long-range attacks

The finality of Ethereum 2.0 is resolved by the Casper protocol.

There are three types of PoS-type consensus: Nakamoto-type consensus, PBFT-based consensus (Tendermint, Casper FFG) and chain-based consensus (Casper CBC). But only the latter two can achieve finality.

The so-called finality means that when a block is confirmed, it will basically not be reversed. The consensus of Satoshi Nakamoto relies on probability to "determine" finality, that is, it can also be overturned by small probability events. The finality of PBFT or CBC is the finality of 100% probability, unless more than 1/4 (CBC) or 1/3 (PBFT) of the verifiers stand up to object, the result can be reversed.

Casper FFG has improved PBFT, inherited the advantages of PBFT, and designed for the irresistible collusion of PBFT, and added the design of abstract consensus mechanism, anti-long-range attack, catastrophic collapse and other mechanisms, and began to integrate fragmentation The scheme becomes the consensus mechanism of ETH2.0 phase0.

Phase0 will start to adopt a modified FFG, and finally complete the conversion of CBC in Phase2 or later. CBC has higher security and theoretical properties, but its complexity and efficiency are low, and it is difficult to implement.

Fragmentation is the basic design of 2.0 to achieve expansion

Why sharding? Changing the consensus protocol to PoS does not improve scalability, which relies on sharding. Sharding is a computer term, and sharding of databases has been around for a long time. The difficulty of fragmentation lies in maintaining security, because once fragmented, the security of each fragment must be maintained by itself.

The main method of 2.0 is to randomly select a group of verifiers by shuffling Shuffle and pseudo-randomly to vote for the shard chain. If there is no randomness, then a large number of malicious verifiers may be assigned to a random verification committee, so that through shuffling, it can be determined that even if there is 1/3 of malicious attackers, the probability of successful attack on the shard chain Still low.

In addition to sharding, Layer 2 has always been the main way to scale. Layer2 is still necessary under 2.0. At present, if sharding can be implemented smoothly under 2.0, the necessity of Layer 2 may not be great. But the key lies in the timing and stability of the sharding implementation.

Vitalik himself is very active in the Layer 2 solution represented by rollup. The realization of sharding will take several years. At present, Layer 2 can be used directly on 1.0, and the expansion will need to rely on Layer 2 in the next few years, so this is very important. A direction is necessary. Even if 2.0 sharding is implemented, Layer 2 may become less necessary, but it is still an optional solution, and it will be very mature compared to sharding at that time.

Validator Responsibilities

The verifier validator has three tasks in Phase 0:

• Validate the attesting beacon chain every epoch;

• Collect attestations of validators on the same committee;

• After being randomly selected into the validator team, blocks will be generated on the beacon chain.

The time interval in Ethereum has two units: a slot is 12 seconds, and an epoch is 32 slots=6.4 minutes. In one epoch, the verifier will produce blocks in the form of a verifier committee, and each verifier committee will appoint a slot.

The verifiers in the committee verify the head of the beacon chain and broadcast it. The block proposer will be randomly selected from the committee. The randomly selected verifiers will be responsible for producing blocks and receiving rewards. If they violate the rules of producing blocks (such as secondary voting or round-robin voting), or are offline, they will be fined for slashing.

We put the discussion on slashing in the next section.

The Staking Problem of Ethereum 2.0

The most important feature of Ethereum 2.0 for participants is the possibility of staking, and it can start from Phase 0, which involves many issues such as pledge returns, long-term lock-ups, economic rewards and punishments, etc. This part attempts to discuss important aspects and answer.

Beacon chain pledge participation rate

As long as 32 Ethers can be migrated to the Beacon chain limit, there are already nearly 117,000 addresses with more than 32 Ethers. Although the number has been increasing, judging from the slope, the rate of increase has remained unchanged for two years, not because of the recent news of the launch of Ethereum 2.0.

Because the transfer is voluntary, the user is nothing more than to consider a few things:

• lock-up yield;

• Difficulty of operation (reward and punishment mechanism);

• liquidity risk;

• Opportunity risk.

The rate of lock-up rate is not completely determined, but is finally determined according to the number of participants. Vitalik gave a calculation of the rate of lock-up rate, which is roughly as follows:

If the minimum number of people staking is exactly 524,288 Ether (or 16,384 accounts holding at least 32 Ether), the yield is very high. But this seems to be very difficult to do. As shown above, there are as many as 117,000 addresses with more than 32 Ether. Only in this way, as many as 3,740,000 ethers need to be mortgaged.

Supplement: Phase 1 of Ethereum 2.0 requires at least 262,144 accounts or 8,388,608 ethers to participate

Of course, there will be many people who take a wait-and-see attitude, but there will also be many users with more than 32ether in a single address. Many large households hold a very large amount of Ether, and users can allocate multiple accounts to participate in mortgages, so we expect pledged Ether to exceed 3 million It's not difficult, and the rate of return is likely to drop below 10% (according to Vitalik's calculation table).

In addition, those who want to participate in the pledge service with less than 32 can pledge through the staking service provider's service similar to "merging accounts", so the final pledge rate will not be too low. Consensys conducted a survey. Those users who are willing to be validators expect the number of nodes they run. It can be seen that it is very common for a user to run multiple nodes.

Difficulty of Operation - Reward and Punishment Mechanism

Ethereum 2.0 introduced a penalty mechanism, which is further subdivided into two categories: slashing (slashing) and inactivity leak (not as a penalty).

Inactivity leak requires the validator to be online continuously, otherwise it will lead to inaction penalty. If the verifier has double voting or round-robin voting, he may be fined. If the fine is lower than 16eth, he will be kicked out of the verifier team.

Therefore, the penalty mechanism is the most concerned issue for staking, which also makes many users choose the staking provider as the next best thing, but even so, they are still most concerned about slashing.

The three functions that users are most concerned about: 1) non-custodial wallet; 2) penalty mechanism; 3) compound rate of return

Because of the existence of punishment and slashing mechanisms, ETH2.0 also provides corresponding positive rewards. The following is an introduction:

reverse penalty

Do not confiscate inactive leak online:

Offline slashing is mainly for the situation that the verifier is not online. There are basically two modes:

1) simply missed the verification of some blocks, so that you just lose the block rewards you deserve, and normal things such as offline maintenance are acceptable;

2) In addition, Ethereum 2.0 requires all 2/3 verifiers (funds) to be online to ensure the security of the entire network. So when there are large-scale verifiers offline (for example, more than 1/3), the system will automatically reduce the fund balance of offline verifiers until the fund balance of online verifiers exceeds 2/3 of the entire network. This situation may happen, but the probability is not high, such as the situation of a world disaster;

3) There is also an extreme situation, which is offline and exit-join, which may trigger the same punishment as the first slashing of Cosmos, that is, two verifiers respectively verify different transactions or blocks. Although they are not intentional, they still encounter To the slashing of the system.

Penalty slashing:

Repeat the double target vote. The extreme case of repeated voting can become a double-spend problem on PoW-type public chains. It mainly means that the verifiers voted for different blocks in one epoch, which will cause two blocks to be produced at once. If a validator uses one Ether to transact twice separately, and then votes for two blocks separately, this will become a double-spending problem. However, the prevention of secondary voting is not mainly for the double spending problem, but to prevent being attacked.

Repeat proposal double proposal. A duplicate proposal means that the proposer proposes two different blocks in one slot.

Positive motivation

Positive motivation

The positive rewards are also divided into three types, the first two are related to slashing, and the third is divided into five categories, which are general validator rewards:

Whistleblower Rewards:

When a validator can provide other validators violations (this behavior is also judged as slashing behavior)

Rewards for block creators:

Those verifiers will be randomly selected as proposers to receive block rewards. Ways to get rewards in Excellent:

• Include evidence of a valid report;

• Join new attestations of other validators. The reported rewards will be distributed to the reporter and the block creator according to 7/8 and 1/8 respectively, but in Phase 0, all rewards will only be distributed to the block creator.

reward for reporting

Proving rewards:

The prover reward means that a verifier agrees to a resolution in 2.0, and such rewards are also divided into five categories:

• contains all validations;

• specify the correct checkpoint validation;

• Enact correct blockchain header verification;

• Fast verification on the chain;

• Point to the correct block in the given shard.

These five types of rewards are all linked to the base reward. The calculation method of the base reward is as follows:

Among them, the effective balance is the effective mortgage balance of each verifier, the basic reward factor is a constant, the total balance is the sum of the effective balances of all verifiers, and the basic reward of a single epoch is also a constant and reflects the sum of rewards within an epoch. The specific calculation of each reward can participate in the CodeFi report.

Why is punishment and forfeiture necessary?

Slashing is a controversial setting in ETH2.0, but the Ethereum Foundation also clearly mentioned when explaining CBC Capser that the significance of enabling Slashing lies in the following two points:

Defend against attack. Slashing makes 51% attacks costly and quick to recover from. This increases the cost of the attacker to more than the gain, allowing the attacker to choose to be a verifier.

Customer Service Verifier Dilemma. The verifier's inaction is the biggest problem in the verifier's dilemma, and the penalty for signing wrong items can be effectively alleviated.

Risks and solutions of long-term lock-up

The long-term and uncertain Ether lock-up is an important reason that prevents users from staking ether. In Phase 0 of 2.0, the pledge of the Beacon chain is a one-way irreversible process. The 32 Ether pledged by the user will not be able to be traded. Although there can be compensation for staking, the user cannot freely transfer money, so this will also hinder the public The willingness to transfer to the Beacon chain, because it is related to whether the launch of Beacon can start with a minimum of 16,324 validators or a mortgage of 524,288 ethers.

Consensys research mentioned that organizations like the liquidity staking working group are currently researching the subject of liquidity staking, which can be applied to Beacon chain staking, but will face the separability of value, pledge/liquidity ratio, and validator conversion Issues such as risks are worth exploring.

Liquidity staking means that users can obtain the liquidity of staking assets while staking, and can use staking as a pledged asset. There will be two types of pledges for liquidity, one is centralized, including mining pools, wallets or exchanges, etc., through a large number of pledges to complete liquidity switching, users can directly complete liquidity management and trade. The other is the way of decentralization, which is completed through smart contracts, that is, it can only be realized in Phase 2. So at present, the centralized approach has a great first-mover advantage.

Of course, if the centralized mining pool or the exchange becomes bigger, it will be more problematic if it is too centralized. For example, the exchange uses pledged Token to vote in the direction opposite to the wishes of some contract users, and even large mining pools jointly bribe elections, etc. (to Eth2. 0 in the CBC stage, there is an anti-bribery mechanism). On the contrary, the decentralized solution is based on Eth2.0, which is part of the ecosystem and may be supported by the community. It just depends on whether users choose economic interests or governance power. At present, most users are more concerned about the economy (this is also the current dilemma of blockchain governance), but it does not mean that factors at the governance level will not be important in the future.

We understand that there is basically only one way to solve the liquidity lockup at present, that is, when Ether is locked, the custodian (it can be a validator or other partners) simultaneously issues a liquidity Token, and the Ethereum The rights and interests of mining are assigned to this Token, and this Token can also be circulated. Because there will be a long BETH lock-up period, this newly issued Token will be active for a long time. It not only has futures attributes, but also has certain bond attributes, similar to a 2-year-maturity treasury bond futures (based on BETH’s super In terms of credit), it will become a big market.

In terms of the comprehensiveness of services, centralized services are still the most likely to escape. It will involve Token issuance, circulation, recycling, payment and staking services, so centralized exchanges, wallets or custodians are most likely to integrate these services, which is a considerable new share.

There are also some decentralized projects that have begun to provide 2.0 corresponding liquidity services, such as Rocketpool, which issue tokens representing BETH through smart contracts to connect users and node operators. Equivalent to a liquidity plug-in based on smart contracts, both users and staking providers can take over and get Token rewards. However, since smart contracts can only be implemented in 2.0, the market rhythm is not dominant.

The business implications and opportunities of Ethereum 2.0

The launch of Ethereum 2.0 is a positive thing for the entire blockchain ecosystem:

First, Ethereum 2.0 will achieve higher security and availability (theoretical and practical levels), proving that the blockchain can be used as a world computer to carry applications. Once the business model of Ethereum proves to be mature, the entire blockchain will enter the mainstream commercial society, and VC/PE financing will accelerate.

Second, I haven't seen any plans after Ethereum reaches 2.0. Tranquility may be the final stage. Because of the long development time cycle, it provides a detailed reference sample for the entire blockchain, and new chains that start later can continue to learn from experience.

Third, the interoperable chain can directly interact with Ethereum, making large-scale asset migration and application migration possible.

Fourth, shard chains are more inclusive, and various chains can also choose to become part of the Ethereum ecosystem.

For Ethereum itself, more commercial implementations can be achieved:

First, DeFi will be able to compete with CeFi at the user level. At present, the performance of DeFi in all aspects is still impossible to compare with CeFi. The reason is that the bottom layer is too slow, which makes it unfavorable to deal with extreme events, and the ease of use has not yet reached the threshold of Internet products.

Second, Dapps that have been popular before can re-enter the stage, and even re-establish exclusive application chains from the perspective of sharding chains.

Third, Ethereum 2.0 can become the best choice for asset on-chain, an absolutely safe and universal platform. It is expected that the on-chain assets carried on Ethereum will explode, with a high probability of exceeding the market value of Ethereum itself.

Fourth, the state of Ethereum becomes valuable, and new leasing business models may emerge.

Although there is no specific promising direction at this stage, what is truly imaginative is that when multi-category, high-value assets start to operate on the chain, the ecosystem will begin to change quantitatively and lead to qualitative changes, such as AMM-type transactions The market may take more share from centralized exchanges.

Another question to consider is whether the interoperability of the blockchain will cause a siphon effect. For example, the assets on the chain are concentrated on Ethereum through cross-chain, which may have a greater impact on some second-tier public chains. The actual reflection is that the market prefers more liquid mapped assets rather than actual assets themselves, because mapped assets are in a safer, more decentralized, and more liquid market, and Ethereum 2.0 may become a multi-asset class trading market.