作者：Blockworks Research，Westie
Compiled by: TechFlow
Given the popularity of the EVM among developers, there has been a desire to combine the scalability of zk Rollup technology with the developer experience of the EVM. This has led to an arms race for zkEVM systems, with the main players being Scroll, zkSync, and Polygon, with Scroll remaining relatively low-profile relative to the other two.
Due to the limitations of the EVM, the zkEVM faces a necessary tradeoff between compatibility/equivalence and performance. Vitalik Buterin listed a rating from 1 to 4 in an effort to clarify the differences between the various zkEVM design choices:
Scroll is in the middle of this tradeoff: they are building a 2.5 zkEVM.
This means they are equivalent to the EVM, with only very minor changes to the way gas costs are structured.
What makes Scroll different
By maintaining EVM equivalence, builders are able to migrate their applications directly from Ethereum L1 to Scroll without making any modifications to their code. They are also fully open source, allowing anyone to use and verify their code.
Scroll also uses a fork of Geth for their sorter, allowing it to inherit a battle-tested system while achieving optimal compatibility with Ethereum.
This also means they can easily apply any new upgrades to Geth; other zkEVMs may require increased code complexity to upgrade.
Scroll Architecture
Scroll's architecture has three main components: Scroll nodes, Roller network, and validation bridge.
The purpose of a Scroll node is to build blocks from users’ transactions and submit them to L1, as well as receive messages from L1.
The three main components of the Scroll Node are the sequencer, the coordinator, and the repeater.
The sorter receives a batch of transactions from the L2 mempool and executes them to generate a new L2 block.
The execution trace of the block is then sent to the coordinator, which passes it to a randomly selected Roller to generate a proof.
Relayers monitor the status of L2 blocks, their data availability and validity proofs from rolling contracts, and deposit and withdrawal events from bridge contracts to ensure everything can be verified.
The Roller network is responsible for generating the validity proof. The Roller network will receive the execution trace from the coordinator, send it to seven different proof circuits, aggregate it into a final circuit, and send the final block proof back to the coordinator.
The verification bridge consists of two contracts: the Rollup contract and the cross-chain bridge contract.
The cross-chain bridge contract handles messages sent between L1 and L2, including bridge transactions of ERC-20 Tokens.
The Rollup contract receives and stores the state root and block data as Ethereum state and calldata. Once a block proof is verified by the Rollup contract, the block is considered final.
In summary, after the sequencer creates a block sequence and generates its proof, multiple block proofs are aggregated into a single proof and sent to the Rollup contract through the coordinator, where the block is verified and finalized.
route map
For the past 3 months, Scroll has been in Pre-Alpha testing, including permissioned smart contract deployment and whitelisted user access.
The next phase will be the Alpha testnet, where they will allow permissionless smart contract deployment that anyone can use.
Scroll’s mainnet will have a two-phase launch.
The first phase will run using a centralized sequencer and decentralized provers;
The main goal of the second phase is to decentralize the sorter.
While Phase 2 is still under active research, current plans will include a PBS model for MEV withdrawals, an on-chain challenge mechanism, and a committee of sequencers to allow for finality before transactions are finalized on Ethereum L1.