Astar 백서

Astar

원문

bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 3쪽 중 2쪽

define T otalIssueRatio ​ , which is the sum of ​ IssueRation ​ . Also, ​ α ​ 1 is the amount of PLM issued per unit ​ IssueRatio in the first Lockdrop. This is an important value to determine the amount of PLM issued in the second and subsequent Lockdrops. Define the number of PLM issues per unit ​ IssueRatio for the second and third times to satisfy ​ α ​ 2 and ​ α ​ 3 ​ ​ the following equation. α ​ 1 ​ : ​ α ​ 2 ​ : ​ α ​ 3 ​ =6:5:4 From the above, the amount of PLM distributed to the second and third user ​ i ​ is as follows. α ​ j ​ × IssueRatio ​ i ( ​ j ​ = 2 ​ , ​ 3) This allows the user to get the amounts of tokens proportional to ​ IssueRatio on the second and subsequent Lockdrops. This solves the problem that if a large number of users do Lockdrop after the first Lockdrop, the amount of PLM that users can get will be excessively small relative to the overall ratio. The following figure6 shows an example of how the amount of tokens distribution changes in multiple Lockdrops. Here, ​ DollarRate ​ is fixed. Figure 6: Multi Lockdrop example. 6.2 Why issue Lockdrop tokens? • We do not hold the assets of the Lockdrop user. • Users do not need to consider the risk of stealing assets by scammers. Issue a PLM with the opportunity cost as collateral. Assets locked by the user will return after the Lock period has expired. • Users can join Lockdrop at a low cost. Anyone who can run a smart contract can participate in Lockdrop. All token holders have the opportunity to participate. • Unlike Airdrop, Lockdrop participants pay a cost to get a PLM. You can issue tokens with a non-zero value. 6.3 Why split Lockdrop multiple times? • If all PLMs are issued on the first Lockdrop, a small number of token holders may own a huge amount of PLM. If you do, there is a risk that a healthy ecosystem will not work. Prevent excessive first-mover benefits. • Users can increase their chances of acquiring tokens by performing multiple Lockdrops. Allow more people to earn a PLM. 7 Real-time Lockdrop Real-Time Lockdrop is a mechanism for 2-nd and 3-rd Lockdrop in Multi-Lockdrop described in the previous chapter. In 1-st Lockdrop, after the period, tokens are issued at once in the Genesis block. Real-Time Lockdrop allows you to get a PLM token immediately after you lock during the Lockdrop period. The details are ​ here. ​ ​ https://docs.plasmnet.io/workshop-and-tutorial/real-time-lockdrop 8 Lockdrop During Polkadot Auction Becoming a Polkadot Parachain is our highest priority. Plasm Network comes into one’s own when it becomes a Parachain because the network can borrow shared security from Polkadot and get interoperability among other Parachains through Inter Chain Message Passing (XCMP)[24] mechanism. The DOT lockdrop will be di ff erent from other tokens. First, we will be starting the DOT lockdrop during the Polkadot Parachain auction. Instead of using a smart contract (or its equivalent) to lock the coins, we will ask the participants locking in DOT to lock it on our Parachain slot. Once the middleware confirms that the DOTs are locked, we can calculate the appropriate PLMs and send them accordingly. This means we are able to issue tokens for DOT lockers at a better rate. This way, we can honor the participants ​ ’ trust in us by becoming a Parachain and providing the participants with the native token for our platform, making it a win-win situation for everyone. Polkadot auctions Lockdrop uses Parachain deposit. Locked DOT will be used to deposit Plasm Network into Parachain. The lock period of DOT expects about two years, during which Plasm Network operates as Parachain. Keep in mind that this Lockdrop can fail because the decision to join Parachain is made at auction. If unsuccessful, DOT will be returned without being locked and PLM will not be got. If successful, DOT is locked and you can get PLM tokens. Figure7 shows the procedure of Lockdrop in Polkadot. Figure 7: Polkadot auctions Lockdrop. This system uses the crowdfund module[25]. Also, note that the DOT will not be available during the auction too and it may fail. Because of this, there is a certain risk compared to other Lockdrops, and the LockBonus of Lockdrop by DOT is the highest. Refer to Multi-lockdrop in the previous sections6. Plasm Network’s Lockdrop cost design is based on the cost of DOT Lockdrop. The cost performance of Lockdrop can be calculated by benchmarking DOT Lockdrop and DOT Staking which is a trade-o ff relationship. 9 Lockdrop A ffi liation Program The Lockdrop A ffi liate Program is a program made to incentivize those who have shared any information regarding Plasm Network to their peers. Via the a ffi liate program, participants of the Lockdrop will be able to receive PLM tokens with a bonus rate. In this section, we will discuss the mechanism of the 1st Lockdrop a ffi liation program. The a ffi liate program mainly has these three rules: • Any participants can reference their introducer ​ ’ ​ s Ethereum public address (this is optional for the lockdrop) • Given that the referenced introducer ​ ’ ​ s address is valid, the token issuing rate for the address being referenced will gain an additional 1% of the participant ​ ’ ​ s issuing PLMs. • Any participants who have referenced a valid introducer will receive another 1% increase in their initial PLMs. The PLM given as a bonus is allocated from the tokens held by the community of Plasm Network (15% of the total). The method of becoming a valid introducer is released in the Plasm Network ​ ’ ​ s Discord server. Furthermore, the valid introducers for the 2nd and 3rd lockdrop will be pooled from the participants of the 1st lockdrop. 10 Consensus Algorithm The Plasm Network changes consensus algorithms and reward designs step by step to maintain security. Specifically, it initially takes the form of a Proof of Authority that is run solely by Validators selected by the community. Next, we will change to rewards centered on collator [?] to participate as Parachain. Eventually, we will move to NPoS, which is also used by Polkadot’s Relaychain. Please refer to the PLM Token Ecosystem chapter for details on how to distribute rewards. 10.1 Proof of Authority Proof of Authority is a consensus-building algorithm that operates only with a validator selected by the community. Public blockchains can be vulnerable when few are launching reliable validators. For this reason, PoA will be operated until a su ffi cient distribution of token holders and the existence of potential validators can be confirmed. The validator at this time will be paid according to the specified parameters as in the case of PoS. 10.2 Incentives of Collator Incentives of Collator is an incentive design for operating Collator in Parachain. The Plasm Network is expected to become the Parachain of Polkadot. Parachain needs a Collator to collect transactions, monitor the transactions in the block, and send the block certificate to the Relaychain validator. Collators must be a full node, so you need an incentive to join Plasm Network as a collator. At first, the Plasm Network Collator is selected by the community, as in the PoA mentioned above. After that, it will be changed to be elected by NPoS described later. In other words, the node that was acting as a validator function as a Collator while the Plasm Network is a Parachain. 10.3 Nominated Proof of Staking Plasm Network uses NPoS which is used on the Polkadot relay chain. This consensus algorithm consists of 3 steps. 1. A Nominator selects a validator NPoS[27]. 2. A validator verifies transactions and makes a new block BABE[28]. 3. Finalize the block that was delivered in the network GRANDPA[29]. The block reward is distributed to the validator who created the block and his Nominator. In addition to that, the reward is also paid to the Plasm Network contributors as follows. 11 Dapps Rewards Dapps Rewards is a mechanism that rewards developers or administrators of smart contracts on a rolling basis. The figure8 is an overview of Dapps Rewards. 50% of Plasm Network’s Staking reward goes to application developers who have enhanced the value of Plasm Network. The Plasm Network allows you to assign a smart contract administrator to a smart contract, and this administrator is called an "Operator". The user can also take smart contracts, this action is called “Nominate”, and the person who does it is called Dapps Nominator. As shown below, the operator of the smart contract receiving many nominees can receive the newly issued PLM token from the chain. Figure 8: Dapps Rewards We will define how to distribute this reward to both Operators and Nominators respectively. Define the following variables: • Rewards ​ nominate ​ ​ : The total rewards allocated to Nominator. • Rewards ​ contract ​ ​ : The total rewards allocated to smart contracts. • Rewards ​ nominate ​ i,j ​ ​ : The rewards allocated to the j-th Nomimate fro the i-th smart contract. • Rewards ​ contract ​ i ​ ​ : The rewards allocated to the operator of the i-th smart contract. • n ​ : The number of smart contracts. • m ​ i ​ ​ : The number of Nominate against the i-th smart contract. • stake ​ i,j ​ ​ : The amount of PLM staked by the j-th Nominate for the i-th smart contract. Then, ​ N ominate ​ i,j ​ gives the following reward for this stake. The nominator can get a reward proportional to the ratio of your stake amount to the total stake amount for the smart contract regardless of the smart contract selected. The operator of ​ contract ​ i ​ who received Stake will get the following reward. On the other hand, the operator can get a reward proportional to the ratio of the stake in the smart contract owned by oneself to the stake of the smart contract. This creates an incentive for the nominator to stake on smart contracts that would simply increase the value of the token. Operators can also receive semi-permanent rewards by receiving stakes on smart contracts managed by themselves. We hope this will be an innovative solution to the di ffi cult problem of monetizing application developers (administrators) on the chain. Note: The operators and nominators have to wait to receive rewards. 12 Operator Trading Operator Trading is a mechanism that involves the buying and selling of Plasma applications by the operators. Since the right is tradable, this is similar to M&A. With the above Dapps Rewards mechanism, Operators can always benefit. If you cannot manage the Operator better, it is a good idea to give the Operator to another party. Of course, Operators are not always bought and sold. Operators will give their rights to opponents who give a value that seems reasonable to the value given to them. Those who have been granted the rights of the operators can receive the reward of operators. There is no need to actually switch operations at this time. However, the side that sold the Operator has already lost the incentive to operate the operator in good faith, and the new owner of the Operator will be transferred and will inevitably operate the Operator. Through this mechanism, we assume that the new o ff -chain market will be created. 13 PLM Token Ecosystem Plasm Network’s token ecosystem refers to Polkadot. Therefore, this document includes the same formula and values as Polkadot. The token name of Plasm Network is PLM that is pronounced "PLUM". PLM has four main roles: 1. Staking for consensus, rewards for validators, and nominators. 2. Transaction Fee used to prevent harmful behaviors. 3. Block rewards for Dapps operator, sustainable reward designed for applications. 4. Good / Bad Voting for Dapps Operator. PLM is intended to be used as a liquidity token. Tokens are issued through multiple Lockdrops to prevent zero-value collateral and increase the number of token holders. PLM tokens are expected to be operated at the ratio of 1 : 1 = ​ Staking ​ : ​ Liquidity ​ . 1 : 1 = ​ Staking ​ : ​ Liquidity 13.1 Inflation Model In the previous chapter, we defined the algorithm that determines the issue amount and distribution method when issuing new Plasm Network tokens. The Plasm Network is structured that the new token issuance fee is shared with Dapps Rewards and a reward for securing the chain. The consensus algorithm of the Plasm Network is expected to be NPoS. Thereby, there are two types of Staking actions: Staking (NPoS) for Validator and Staking (Dapps Rewards) for smart contracts. Both rewards from each of the staking are equally proportional to the amount of staking. Users who stake on validators / smart contracts are collectively called nominators. The ideal ratio of Staking for validators and Staking for smart contracts is: Staking ​ validators ​ represents the action of staking on validators. Staking ​ contracts ​ represents the action of staking on smart contracts. Then, the below formula is the expected ratio between Staking for validators and Staking for smart contracts. 5 : 1 = ​ ​ Staking ​ validators ​ : ​ ​ Staking ​ contracts We consider the rewards paid to operators in dapps rewards. Operator rewards increase in proportion to the inflation rate due to staking. Dapps rewards 50% of the total reward when meeting the ideal ​ q from quote the Dapps Rewards chapter. The rewards obtained by the Operator at that time is maximized. To show the specific reward distribution, we introduce the following variables: • Rewards ​ operators ​ ​ are the total amount of reward gotten by the Operator. • Rewards ​ stakers ​ validators ​ are ​ the total amount of rewards gotten by staking a validator. • Rewards ​ stakers ​ contracts ​ are ​ the total amount of reward gotten by staking smart contracts. • t ​ is a coe ffi cient that represents how many times the total amount of rewards earned by the operator is greater than the rewards earned by taking a smart contract. t ​ = 4 ​ from quote the Dapps Rewards chapter and 50% of the total reward for meeting the ideal q will go to the Dapps Rewards reward. Therefore, the ideal distribution ratio of remuneration is determined as follows. Rewards ​ stakers ​ validators ​ ​ : ​ ​ Rewards ​ stakers ​ contracts ​ ​ : ​ ​ Rewards ​ operators ​ = 5 : 1 : 4 ​ t ​ ​ = 4 Also, the percentage of Staking and the percentage of reward are equal as follows: Staking ​ validators ​ : ​ ​ Staking ​ contracts ​ = ​ ​ Rewards ​ stakers ​ validators ​ ​ : ​ ​ Rewards ​ stakers ​ contracts PLM tokens use the same NPoS as Polkadot. This nominator and validator can operate the token at a certain annual interest rate for Staking. Also, token rewards will be paid to the PLM Dapps operator’s Nominator and Operator as well. Plasm Network’s inflation model is defined as follows: First, follow the Polkadot inflation model and define the following variables: • x ​ is the total amount of staking divided by the total amount of tokens issued. • X ​ ideal ​ ​ is the ideal value of ​ x ​ . ​ Staking ​ : ​ Liquidity ​ = 1 : 1, so ​ X ​ i ​ deal ​ = 0 ​ . ​ 5. • q ​ is the amount of staking to the validator divided by the total amount of staking. Staking ​ validators q ​ ​ = ​ Staking ​ validators ​ ​ + ​ Staking ​ contracts • Q ​ ideal ​ ​ is the ideal value of ​ q ​ . From ​ ​ 5 : 1 = ​ Staking ​ validators ​ ​ : ​ Staking ​ contracts ​ , the ideal ​ ​ value of ​ q is ​ Q ​ ideal ​ = 5 ​ / ​ 6. • i ​ ( ​ x, q ​ ) ​ is the average annual interest getting by Staker. It is a monotonically decreasing ​ function of ​ x ​ , ​ |Q ​ ideal ​ − ​ q| (di ff erence from the ideal ratio). To make both ​ x ​ , ​ q close to the ideal value, when ​ x ​ , ​ |Q ​ ideal ​ − ​ q| is low, raise the interest rate as an incentive to increase the amount of stake. When ​ x ​ , ​ |Q ​ ideal ​ ​ − ​ ​ q| ​ is high, lower interest rates as an incentive to reduce stake. • i ​ ideal ​ ​ is the average annual interest rate of Staker ​ i ​ ( ​ x, q ​ ) ​ ​ when both ​ x ​ and ​ q ​ are ideal values. in other words, ​ i ​ ideal ​ = ​ i ​ ( ​ X ​ ideal ​ , Q ​ ideal ​ ). • I ​ Staking ​ is the inflation rate by Staking. I ​ Staking ​ , a bivariate function involving x ​ and q ​ , ​ draws a three-dimensional convex function. Expressing Staking total amount ​ x interest rate = inflation rate and expressing it as ​ x ​ ∗ ​ i ​ ( ​ x, q ​ ) = ​ I ​ Staking Also, this value is maximized when ​ x ​ , q ​ is the ideal value from the reward design of ​ i ​ ( ​ x, q ​ ). The ideal state equation can be expressed as ​ X ​ ideal ​ ​ ∗ ​ ​ i ​ ( ​ X ​ ideal ​ , Q ​ ideal ​ ) = ​ M axmiumI ​ Staking ​ . • I ​ 0 ​ is the lower limit of the inflation rate. When x ​ = 1 ​ or x ​ = 0, converge to the lower limit. I ​ 0 ​ is equivalent to the operating cost of the validator. The reason is that if you do not ​ secure at least the incentive to operate the validator, the chain will break, so ​ I ​ 0 ​ = 0 ​ . ​ 025 is recommended here. • d ​ is an adjustable decay rate for each ​ x ​ . Each time ​ x ​ is ​ d ​ more than ​ X ​ ideal ​ , ​ I ​ Staking ​ ​ is reduced by 50%. In other words, ​ I ​ Staking ​ ( ​ X ​ ideal ​ + ​ d, Q ​ ideal ​ ) ​ ≥ ​ ​ I ​ Staking ​ / ​ 2.We recommend d ​ = 0 ​ . ​ 02. • g ​ is an adjustable decay rate on ​ q ​ . Each time ​ q ​ is ​ g ​ away from ​ Q ​ ideal ​ , ​ I ​ Staking ​ ​ is reduced by 50%. In other words, ​ I ​ Staking ​ ( ​ X ​ ideal ​ , Q ​ ideal ​ ​ pme ​ ) ​ ≥ ​ ​ I ​ Staking ​ / ​ 2. We recommend ​ g ​ = 0 ​ . ​ 15. • i ​ staking ​ is the average annual interest earned by the nominator through Staking. This ​ can be determined by dividing inflation by the Staking ratio. In other words, • I ​ operators ​ is the inflation rate due to the rewards that the Operator can get. This is t ​ times ​ the ratio (1 ​ − ​ ​ q ​ ) of Staking to Operator in ​ I ​ Staking ​ . From the below auxiliary formula1 and formula2, ​ I ​ operator ​ = ​ t ​ (1 ​ − ​ ​ q ​ ) ​ I ​ Staking ​ . Here, the ratio of the amount of reward and the ratio of the inflation rate are equal from the formula above. This represents the average (based on the amount staked) interest rate of the operator's reward. From the auxiliary formula I ​ is the overall inflation rate. This is , ​ which is the sum of the reward for Staking and the inflation rate due to the reward for Operator. The below figure is a graph simulating the inflation rate when each parameter is set as follows. i ​ ideal = 0 ​ . ​ 2 X ​ ideal = 0 ​ . ​ 5 Q ​ ideal = 5 ​ / ​ 6 I ​ 0 = 0 ​ . ​ 025 d = 0 ​ . ​ 02 g = 0 ​ . ​ 15 t = 4 Figure 11: The inflation graph of ​ q ​ = ​ Q ​ ideal ​ .( ​ https://www.desmos.com/calculator/ ​ ​ v8mrxdwbvz ​ ) The figure11 graph is fixed at ​ q = ​ Q ​ ideal ​ . Here, the upper green line is the average annual interest rate ( ​ i ​ operators ​ ) for the operator’s staking amount, the lower green line is the average annual interest rate of the staking ( ​ i ​ staking ​ ), and the red line is the overall inflation rate ( ​ I ​ ), the upper blue line indicates the inflation rate due to Staking reward ( ​ I ​ Staking ​ ), and the lower blue line indicates the inflation rate due to Operator reward ( ​ I ​ Operator ​ ). The inflation rate when both ​ x ​ and q ​ are ideal values is 0.166 ... (1/6) at maximum. Next, the graph when q ​ = 0 ​ . ​ 2 ​ is shown ​ in Figure12. When ​ q = 0 ​ . ​ 2, the Staking percentage is 1 : 5 = ​ Staking ​ validators : ​ Staking ​ contracts ​ , and the reward percentage is as follows: Rewards ​ stakers ​ validators ​ ​ : ​ ​ Rewards ​ stakers ​ contracts ​ ​ : ​ ​ Rewards ​ operators ​ = 1 : 5 : 20 Figure 12: The inflation graph of ​ q ​ = 0 ​ . ​ 2 When the figure12, although the ratio of the operator’s reward is increasing, the upper green line representing the average annual interest rate for t
프로젝트로 돌아가기