Today we continue to talk about projects. As for the leading classification of each track (2), we will talk about it later. Today’s project is also very good. It is also a player on the L2 track. Its performance has been quite strong recently, rising from 18 to 88. The current market value has reached 400 million US dollars, and the ranking has directly rushed to 130+. It is METIS. It is actually an L2 of OP and solves some problems of the current OP L2.
Currently, many popular optimistic rollup second-tier solutions suffer from some important issues:
In layer2, the role of the sequencer is to sort and compress hundreds of transactions into one transaction, and then hand it over to layer1 for confirmation. It can be said that it realizes the high performance and low handling fee of layer2. However, if important things are centralized and there is a risk of evildoing or single point of failure, the current mainstream approach is fraud proof. Even if anyone can verify the authenticity of its transactions, this design leads to the second a question.
It often takes a long time to complete a transaction, especially when it comes to extracting from a second-tier structure to a first-tier structure. It may take several days for the transaction to be completed.
This in turn will again lead to bottlenecks on the second floor. Due to the way many Layer 2 solutions are constructed, Layer 2 throughput will ultimately be limited by the capabilities of the centralized stack. If the goal is to push DApps into the mainstream, second-layer solutions must be allowed to scale.
Merits’ solution-MVM virtual machine
In MVM, each Decentralized Autonomous Company (DAC) is associated with a unique two-tier structure, where all DAC collaboration takes place (the purpose of this design is to allow more companies to participate , rather than operating independently like ARB and OP).
MVM separates the computing and storage built by Ethereum into two layers. More specifically, the creation of a DAC only triggers the creation of a new storage layer specifically for that DAC. The computing part, including block mining and cross-layer communication, will dynamically expand and contract horizontally according to demand as part of the Metis system. This is a key design highlight that allows Metis to scale out without incurring major infrastructure expenses. Revenue from economic activity within each DAC will easily cover infrastructure costs.
此外，MVM 允许提供者注册并贡献计算能力，使二层结构真正去中心化。提供者将根据其产生的区块而获得激励。关键点在于提供者可能不知道将在哪个 DAC 上工作，这有助于确保不会有恶意计算能力提供者的有针对性攻击。
Sequencer is responsible for the connection between Layer 2 and Layer 1 (through the Rollup contract of Layer 1); the storage layer provides data services for Metis virtual machines (MVMs); MVMs execute specific Layer 2 logic; Rangers are responsible for the correctness of random data, in Layer 2 Conduct verification (trying to solve the problem of Layer1’s recognition of the data correctness of Layer2, and quickly extract assets from Layer2).
As part of the MVM design, the first-level contract knows how to handle state roots, transactions, and proposals from different second-level contracts. Each transaction batch and state root is also accompanied by a chainID associated with the payload. chainID is a unique identifier for each layer 2 structure and is also used to sign transactions between different layers.
When a DAC needs to be created, as part of the Comco protocol, a layer of smart contracts will generate a unique chainID for the new structure, register it with the Metis management smart contract, and trigger the Metis microservices to deploy storage units.
DAC Charter to standardize operating rules
Just like organizations in the real economy, DACs have charters that set their basic operating rules. MVM_DACMaster is a smart contract that specifies the operation of DAC on the second layer of MVM. This smart contract is part of the genesis block and contains the following important information about DAC:
˙vision:string and mission:string – the purpose of DAC
˙permissions:{address:string[], operation:string, opcode:string}[] – A set of addresses that can perform specific operations on the second layer of MVM, including basic read and write permissions.
˙taxrates:{threshold:int, rate:int}[] – Mechanism that defines how transactions within DAC are taxed.
˙management_contract:address – Define the process of changing the content of the DAC charter, such as a voting mechanism similar to DAO, or based on smart contract settings that cannot be changed.
The Charter will be created at the creation of the world. The initial settings will be finalized in the first block after genesis as part of the DAC creation process. The management contract will be the only signatory able to amend the charter thereafter. The only exception is the tax rate, which Metis is managed by MetisDAO and can be set.
MVM second layer permission management
In any organization, structure and authority are two important factors that influence effective functioning. As part of MVM, it introduces a flexible permission system that can precisely manage access to specific layer 2 DACs at the opcode level.
In the DAC charter, each opcode is associated with a set of addresses that are allowed or prohibited from executing that opcode during execution. The MVM_ExecutionManager responsible for executing the opcode checks permissions before executing the opcode. Minimize the impact of permission checks during execution by implementing extensive caching. Some examples of permission implementations include, but are not limited to:
˙Only a few signers can deploy smart contracts - this allows DAC to have greater control over its own ecosystem, including token distribution.
˙Only DAC participants are allowed to submit transactions - this configuration can create a permissioned DAC.
˙Some signers are prohibited from submitting transactions - In this way, DAC can blacklist harmful members or malicious users.
A special storage layer based on IPFS is used to manage confidential data.
Another important component is MVM's special storage layer. MVM has a regular storage layer for storing blocks and state. However, for some DACs, there may be some sensitive information that they do not wish to be disclosed outside of the DAC. Since all transactions are packed into L1 for potential dispute scenarios, confidential data on the blockchain can still be accessed outside the DAC. Encryption is often adopted as a solution to this challenge. However, encryption and decryption operations typically require a central service manager that provides access to encryption key pairs. At Metis, it decided to use IPFS to solve this problem.
Ways to Reduce Fraud
In all rolling solutions, fraud prevention is always part of the trade-off. Optimistic rollups typically have a long anti-fraud window to provide enough time for Rangers to verify the state root. This longer window makes withdrawals to L1 less efficient. As part of MVM, its goal is to minimize the fraud prevention window by decentralizing the rolling process and promoting state root verification.
Based on this work, by separating computation and storage, the rolling process will no longer be handled by a single sequence generator. A set of sequence generators will be randomly selected to roll the state root and submit transactions.
Additionally, MVM introduces a unique character called the L2 Ranger. L2 Rangers are members of a special DAC called MVM_RANGERS. Rangers are able to perform block-wide sampling of periodically distributed transactions from random DACs (including MVM_RANGERS itself) and verify the state root against the transactions. Each completed verification will be presented in the form of incentives in Metis Token (MT). If the verification finds a difference, you can withdraw to L1 and the challenge process will start automatically. A successful challenge will reward the validator with a portion of the deposit. Failed challenges will cause the validator to lose their deposit and eventually lose access to MVM_RANGERS. All of the above work is intended to make the arbitration game fairer and more attractive to Rangers. Through decentralized ordering and a healthy L2 Ranger pool, MVM will greatly shorten the required attestation window and further improve network efficiency.
Dynamic sequence generator margin to protect the network
Another important mechanism to prevent malicious sequence generators from committing fraud is security deposits. Each sequence generator needs to mortgage a certain number of Metis Tokens to qualify. However, the Metis ecosystem is closely tied to the real economy where transactions can be worth billions of dollars. The risks and rewards of malicious behavior begin to favor the more dangerous side. Therefore, MVM introduces a concept called Dynamic Margin Threshold (DBT).
DBT is calculated based on the maximum economic capacity of a given DAC, which is the total supply of MVM_Coinbase. If the amount of Metis Token (MT) mortgaged by a specific sequence generator is lower than the DBT of DAC, the sequence generator cannot participate in the DAC sorting work. DBT will be automatically updated when MT deposits and withdrawals are made. If after a deposit operation there is no eligible sequence generator in the sequence generator pool, the operation will be blocked until an eligible sequence generator is found.
Hybrid Rollup
On March 2, 2023, MetisDAO announced that it was building the first-ever hybrid Rollup, combining the Optimistic Rollup architecture with zero-knowledge proof zk. According to the ChainCatcher report, on July 13, the MetisDAO Foundation announced the official release of its new incubated project ZKM. ZKM will be launched on the test network before the end of the year, upgrading the existing Optimistic Rollup to Hybrid Rollup to achieve timely withdrawals (original OP’s Withdrawal period is 7 days) and ensure security.
It is reported that ZKM uses the MIPS instruction set to implement zero-knowledge proof (ZKP) security for all virtual machines (VMs) and applications from the CPU level, and supports a variety of blockchain smart contract engines. The plug-and-play feature allows developers to apply ZKP without changing the code base, reducing adoption costs. Leveraging Ethereum’s decentralized security infrastructure, ZKM can verify all blockchain and non-blockchain transactions. The goal of the project is to expand beyond L2 to L1, cloud computing and the Internet of Things, making Ethereum the settlement layer for the global blockchain and non-blockchain world.
ecological development
Its ecological development has been very rapid recently, and there are currently 81 applications displayed on the official website. DEFI has AAVE, DEX has sushiswap, infrastructure has chainlink and so on.
Token economy
The project was launched in May 2021. The current total number of tokens is 10,000,000, the total circulation is 5,420,000, and 4,573,674 have been circulated, so the current circulation rate is only 45%. The current currency price is 88 US dollars, and the peak was $323.35 (2022-01-17). Although it has increased 5-6 times, there is still a certain gap from the high point. In terms of token distribution, 7% of the founding team + 6% liquidity reserve = 13%, so the team retains very little. Then I checked TVL, and it was about 60 million US dollars, which is not very high. There is a big gap between ARB and OP.
In conclusion, although this project is somewhat suspected of robbing ARB and OP traffic, it seems reasonable. The sorter is centralized, a bit like SSV. I provide you with a decentralized solution for LIDO to use. Of course, this project also has certain technological innovations and certain ideas. A decentralized world requires everything to be decentralized, which is more in line with web3. And it is not a new project, it is also an old project. Why? It has suddenly risen recently. The TVL of the First Rift has grown to a certain extent, which is related to its continuous construction. In the Second December, they used 4.6 million Metis tokens to stimulate ecological construction. Third, the recent Ethereum Kunkan upgrade has caused all L2 to rise. Of course, the fundamentals of this project are actually pretty good, at least not bad. It’s just that the point it cuts into is not a very painful point at the moment.