How to staking in ETH2.0, we asked V God
Author: Rudy Ludi
Yesterday I was honored to be drawn into the newly established Ethereum 2.0 Chinese community. When I entered, there were only a few dozen people in the group. I saw that V God’s WeChat is also in the group, and V God seems to have a lot of time (later I learned that it was in Waiting for the plane at Taipei Airport), has been actively interacting with Ethereum 2.0 enthusiasts in the group (Chinese expression skills are getting stronger and stronger), and the whole afternoon turned into a very long online AMA.
The AMA talked about sharding, governance, DApp, 1.0 and 2.0 transition, etc. My friends and I saw the opportunity and of course asked a lot of staking-related questions. Before that, we could only look for V on the public network such as the Ethereum forum. The fragmented interpretation of God and other people, today I finally got V God's personal explanation. Some questions I wanted to ask before have been answered, and I feel very comfortable.
secondary title
Interpretation
1. 32ETH is the pledge value required to become a node, which is both the lower limit and the upper limit. If there is 320ETH, only 10 nodes can be established, and 320ETH cannot be pledged to 1 node.
2. 32ETH, corresponding to 1 node, corresponding to 1 private key.
3. Multiple nodes can be run through one client, but the number of blocks to be verified and signed will also increase accordingly. If hundreds of nodes are running at the same time, it may be necessary to rent a server.
4. The design of ETH 2.0 is different from that of popular staking chains such as Cosmos and Tezos. In the early stage, there is no concept of "entrustment", that is, token holders cannot decentralize the ETH they hold to nodes for mining .
5. But in phase 2 (phase 0 is the first to be opened), it will be possible to become a node through the contract. At that time, the node service provider can initiate the node contract and set the rebate rules. Participate and get paid.
secondary title
Development Opportunities for Staking Service Providers
According to the current explanation given by Vitalik, decentralized entrustment cannot be realized in phase 0 and phase 1, so the holders of ETH cannot pledge ETH to the node for mining, and it is difficult for Staking service providers to follow the previous establishment of nodes, Community canvassing, rebate dividends and other models, so what opportunities will there be?
Although the purpose of Ethereum's design is to encourage more users to participate in mining and realize the decentralization of nodes, it is still difficult for coin holders to directly participate in the node network of ETH2.0. Firstly, 32ETH is still not a low threshold, which will prevent a large number of small currency holders from participating; secondly, nodes need to run 24 hours to produce blocks, although Ethereum has tried its best to minimize the punishment brought by offline to encourage public participation , but 24-hour operation is also a tiring job, not to mention node upgrades, operation and maintenance, disaster prevention and other troublesome things.
Whether it is the inability to participate due to the small number of coins held, or the insufficient income caused by the inability to be online 24 hours a day, it will cause invisible inflation losses under the additional issuance of ETH, which must be something that ETH currency holders are not happy to see. In response to such demands, Staking service providers can explore several paths for ETH 2.0.
1. Provide technical support services
Let users master the private key, and Staking service providers provide them with complete node technical services, including server leasing, node upgrades, operation and maintenance, disaster prevention, etc., and charge a certain technical service fee.
2. Centralized wallet
Although it is not possible to decentralize the entrustment, it is possible to provide a unified centralized wallet for users to deposit coins, and the Staking service provider will set up a node for every 32 ETH of the deposited coins, and rebate all deposit users according to a certain rebate rate. This method can not only solve the problem of currency holders running nodes, but also allow currency holders with less than 32ETH to obtain Staking income, but the disadvantage is the risk of nodes running away, and the provider of the centralized wallet needs to obtain ETH currency holders full trust.
3. Waiting for Phase 2 to go online to realize decentralized entrustment
secondary title
AMA record
1. Node conditions
Zhufeng: I am concerned about what conditions ETH2.0 needs to be a node
Vitalik Buterin：32 ether and a normal computer should be enough.
Dazhi: What if there are more than 32? What if there are 320 coins? Divide into 10 nodes to run?
Vitalik Buterin：You can run a client, and the client can manage multiple validator ids (validator identities). But the more validator ids, the more data needs to be verified. For example, if you have 10,000 ether, you may need a server.
Kai: The more data mentioned here, does it mean the same data on the chain, the more times a number of validators vote (vote)?
Vitalik Buterin：This is a characteristic of sharding. The blocks of sharding are assigned to the validator id, so if you have many validator ids, you need to verify and sign more blocks.
Dazhi: Do you need to keep 10 private keys for 10 ids? Or does 1 private key only need 1 client to manage 10 ids?
Vitalik Buterin：10 private keys. But the client can generate multiple private keys with one private key.
2. Possibility of decentralized delegation
Rudy@InfPool：It seems that the staking of ETH 2.0 should not contain "delegated"part of it. If you don't have enough 32ETH, you can't delegate it to other nodes in a decentralized way? Perhaps the centralized wallet can build a pool (mining pool), artificially allocate, and set up a Validator for every 32ETH.
Vitalik Buterin：After Phase 2 (Phase 2), you can use the contract as an entrustment, and the holder of the private key will sign blocks for you, etc., but you cannot withdraw (withdraw) to your own account.
Rudy@InfPool：You mean, after Phase 2, how many people can collect 32ETH through one contract to become one Validator?
Vitalik Buterin：Well, let me describe this method more.
When you open a validator, you need to set up two public keys: the signature public key and the withdrawal public key. After Phase 2, it is not necessary to set a withdrawal public key, but a withdrawal contract can also be set. The private key of the signature can be withdrawn. After the withdrawal is completed, the funds in the validator will be given to the address of the withdrawal contract. So when you open a new validator, you can first send, for example, 1 ether, and set up a withdrawal contract. The rules of this contract are, whoever deposits (deposits) and withdraws according to the ratio. For example, you deposit 1 ether, and then 3 people deposit 8, 10, and 13 ether, a total of 32, so the validator becomes active (active). After a period of time, you send a withdraw transaction, and now because the validator’s balance (balance) is 33.6 ETH (plus 5%), 33.6 ETH is given to the contract, and the rules of the contract assign 1.05 to you, 8.4, 10.5 and 13.65 to other participants.
The private key of the signature is yours, and the rules for who can withdraw how much are stipulated in the contract. Others can see the rules of the withdrawal contract of a validator that has not been activated, and if they think it is reasonable, they can send their coins to the validator. This is the delegated method.
Rudy@InfPool：Got it, but wait until phase 2.
I first saw the idea of a contract as a validator in this post on ethresearch. The name of the post is "Decentralized Casper Validator Proposal" (a proposal for a decentralized Casper validator), and I recommend it:
https://ethresear.ch/t/decentralized-casper-validator-proposal/1430
3. 32 ETH threshold
blockpunk@无志: The threshold for 32eth is still very high
Vitalik Buterin：The number 32 is set according to the efficiency of the client. If it is 4 ethers, there will be many validator ids, and the overhead of the chain will be too high. However, the improvement of client efficiency can be used in two directions: one is to reduce the minimum eth of the validator, such as to 8, and increase the number of shards at the same time; the second is not to increase these numbers, so that the client becomes more and more low resource, so that more and more users can run beacon nodes (beacon chain nodes).
4. Hardware requirements
blockpunk@无别: A client with 320,000 ethers, the 10,000 vaildators created will be divided into 1024 shards by random numbers, so this client has to store the data of all shards, right? Such a machine The bandwidth and storage can't sup port it.
Vitalik Buterin：yes. The latest plan is not 1024 shards, but 64 shards and 2048 committees. If your ETH exceeds 60,000, you need to process almost all blocks, so local tyrants may need to use the server.
Rudy@InfPool：Is it possible to rent a server to process all blocks? Going up, the difference between 60,000 and 600,000 should not be big.
Vitalik Buterin：learn.
Rudy@InfPool：learn.
