Original title: "Make Ethereum Great Again"

Original author: Redacted Research

In the blockchain world, Ethereum has always been the first choice for many developers and users with its advantages of decentralization and smart contracts. With the rapid development of blockchain technology, Ethereum faces many challenges, including slow transaction speed, long block time and high transaction fees. In order to solve these problems and further improve the performance of Ethereum, MegaETH came into being.

Does the market need other L2s at present?

MegaETH is a parallel EVM L2 backed by Vitalik Buterin, which aims to achieve:

· 100,000+ transactions per second

· Block time and latency less than 1 millisecond

· Transaction fees less than $0.01

The project recently completed a $20 million seed financing round, aiming to push the performance of Ethereum L2 to the hardware limit and bridge the performance gap between blockchain and traditional cloud computing servers.

Requirements and Challenges of L2

Although 80 Rollup solutions have been put into use and the number will continue to grow in the future, simply increasing the number of chains is not enough to directly solve the scalability problem, which is a conclusion drawn by observing the current popular EVM extension chains. In fact, a more accurate metric for evaluating TPS is gas/second (milli) rather than pure TPS.

For example, the opBNB chain stands out among many existing major L2 solutions and EVM-compatible L1s with its high gas processing speed of 100MGas per second. But even so, it still has a lot of room for improvement compared to the processing power of Web2 servers.

For reference, 100MGas/second is equivalent to 650 Uniswap swaps or 3.7k ERC20 transfers per second.

Limitations and Improvements of EVM

The characteristics of sequential transaction processing in the standard EVM impose significant limitations on any application. Application-specific L2s as dedicated and optimized execution environments or high-performance virtual machines can help overcome these EVM-specific limitations. However, this also comes with some costs. The lack of cross-chain permissionless interoperability undermines DeFi and the composability that Ethereum once prided itself on.

Tradeoffs of High-Performance Chains

The synchronous composability of the overall system remains the main advantage that favors chains like Solana over the current decentralized rollup ecosystems on Ethereum or Celestia.

Unfortunately, high-performance L1 chains like Solana, Aptos, or Sui Network also have their own tradeoffs. Increasing validator requirements to scale throughput leads to decentralization losses, coupled with cryptoeconomics being less secure on L1.

In addition, these emerging ecosystems require developers to learn new programming languages like Rust or Move, which not only poses challenges for the EVM/Solidity-based developer community, but also increases the difficulty of launching the ecosystem.

What is MegaETH?

MegaETH Labs is a validity aggregation that verifies state on Ethereum and publishes data to Eigen DA's L2 with very low fees.

As with any L2, the sorter is responsible for sorting and executing user transactions. MegaETH has only one active sorter at any given time, eliminating consensus overhead during normal execution.

On the other hand, full nodes accept state differences from the sorter over the P2P network and apply them directly to update their local state. This does not re-execute transactions, but instead indirectly verifies blocks using the proofs provided by the prover. A key advantage of this node specialization is the ability to set independent hardware requirements for each type of node. Since this sorter node handles heavy execution work, it is best to run them on high-end servers to improve performance. In contrast, the hardware requirements of full nodes can be kept relatively low because the computational cost of verifying proofs is not expensive. This is in line with Vitalik's philosophy of "scale block verification, not production".

How does the modular DA solution differ from other EVM L2?

The EVM is often blamed as the main reason for the relatively low performance of EVM-based L2. This is part of the reason why more and more altVM L2s like Movement Labs or Eclipse are emerging.

This stereotype is incorrect. Performance measurements show that revm (Rust-based EVM used by MegaETH) can achieve about 14,000 TPS on recent Ethereum blocks during a historical sync setup.

14,000 TPS is definitely enough for most L2s. However, MegaETH wants to go a step further. Traditional EVMs face three main causes of inefficiency, including high-frequency access latency, lack of parallel execution, and excessive demand on handlers.

MegaETH labs chose to extend EVM to the limit on Ethereum, and the node specialization inherent in its design enables sorting nodes to be equipped with a large amount of RAM, which can store the entire EVM world state and state tree in memory. This technology, called in-memory computing, is essential for high-performance data-intensive applications in the Web2 world.

How does MegaETH work?

Node specialization is particularly prominent in the design of MegaETH Labs, especially the design of its sequencer nodes. These nodes are equipped with a large amount of RAM, which enables them to store the entire Ethereum Virtual Machine (EVM) world state and state tree completely in memory. This design is crucial because it greatly improves the state access speed, which is 1000 times faster than traditional SSD-based storage systems. This in-memory computing technology is crucial for applications in Web2 environments that require high performance and process large amounts of data. In short, MegaETH provides blockchain applications with processing speed and efficiency close to traditional servers in this way.

Parallel EVM has become a hot topic, and many teams are focused on porting Block-STM, originally implemented for MoveVM, to the EVM chain. However, this is a complex task and has its limitations.

MegaETH uses a low-latency stream-based block building algorithm, combined with a concurrency control protocol that supports transaction priorities, becoming the first real-time EVM execution engine that can process transactions instantly and publish resulting state changes every 10 milliseconds.

In addition, MegaETH uses just-in-time compilation to transparently translate smart contracts into native machine code on the fly. This completely eliminates the inefficiencies of interpreting EVM bytecode and emulating stack machines from the outset.

Bottlenecks in disk I/O operations

The biggest bottleneck for EVM-compatible blockchains is resource-intensive disk I/O operations. Disk I/O operations involve accessing blockchain states stored on disk, including reading and writing contract storage (using SLOAD and SSTORE opcodes).

Disk I/O operations also include access to account balances and other on-chain data required for contract execution. Efficient disk I/O operations are critical to the performance and scalability of nodes, ensuring that state data is retrieved and updated in a timely manner during transaction processing.

MegaETH solves this problem by introducing a new state tree, designed from scratch to replace the commonly used Merkle Patricia Trie. This new tree minimizes disk I/O, is able to efficiently scale to terabytes of state data, and maintains full EVM compatibility.

Efficient State Sync

Finally, MegaETH leverages an efficient P2P protocol to quickly sync full nodes, ensuring that even regular internet connections can keep pace with processing speeds of up to 100,000 transactions per second.

Making Ethereum Great Again

MegaETH certainly has an interesting narrative as the first modular blockchain to support live EVM execution. High-speed transactions and market-leading low fees, while maintaining EVM compatibility and being able to tap into the largest development pool.

As an L2, MegaETH also has a native bridge to Ethereum L1, seamlessly plugging into L1 liquidity and user base to bootstrap its ecosystem. In addition to Solidity support, this is another major advantage over the existing altVM L1s on the market.

MegaETH’s main competitor is likely to be Monad, which is definitely not an easy opponent to deal with, especially considering its strong community.

MegaETH recently announced that it has successfully completed a $20 million seed round of financing, led by the well-known investment institution Dragonfly Capital. The angel investors participating in this round of financing include industry heavyweights such as Ethereum co-founder Vitalik Buterin, ConsenSys founder Joseph Lubin, and Eigen Layer founder Sreeram Kannan.

Through the above series of technical improvements and innovations, MegaETH is working hard to push Ethereum's performance to a new level and truly "make Ethereum great again."

「Original link」

欢迎加入律动 BlockBeats 官方社群：

Telegram 订阅群：https://t.me/theblockbeats

Telegram 交流群：https://t.me/BlockBeats_App

Twitter 官方账号：https://twitter.com/BlockBeatsAsia