Monad: The new generation of PoS is compatible with the first-layer public chain of EVM. Where will the “Ethereum Killer” go?

以南说

24-01-22 11:02

Read this article in 10 Minutes

AI summary

View the summary

Original Title: "Monad: The First Layer Blockchain with Proof of Stake and Ethereum Virtual Machine Compatibility"

Original Author: Yinan

Project Introduction:

Monad is a new Proof of Stake (PoS) Ethereum Virtual Machine (EVM) compatible Layer 1 blockchain aimed at improving the entire Ethereum ecosystem and serving as a decentralized computing platform.

Monad provides throughput several orders of magnitude higher than any existing smart contract blockchain, supporting 10,000 smart contract transactions per second (tps) through algorithmic improvements to virtual machine design and Byzantine fault tolerance (BFT) consensus mechanism.

Technical Innovation:

Monad client is built with a focus on performance, and is written from scratch using C++ and Rust.

ETH Current Performance Bottlenecks

Ethereum's current execution limit, such as the limit of 1.25 million Gas per second, is conservatively set based on multiple factors:

The content you provided is:

低效的存储访问模式. Translated to English, it is:

Inefficient storage access mode.

Ethereum's state is stored on the blockchain, and due to its distributed nature, it is slower compared to traditional databases. Accessing and modifying this state (e.g. reading or writing smart contracts) can be inefficient, especially if the storage access patterns are not optimized. This inefficiency is a key reason for maintaining a conservative Gas limit, as too much state access in a short period of time can lead to network congestion.

Currently, Ethereum adopts a single-threaded execution mode, which means that transactions are processed one after another. This serial processing may become a bottleneck for system throughput. Parallel execution may increase throughput, but it introduces complexity in transaction sorting and execution consistency, making it a difficult feature to implement and optimize.

Very limited execution budget

The execution and transaction processing of smart contracts require computing resources. Consensus mechanisms (especially in proof-of-work and to some extent in proof-of-stake) require nodes to reach agreement on the state of the network. This process cannot continue before execution because consensus is the foundation of normal network operation.

Concerns about state growth and the impact of state growth on future state access costs

With the increasing use of the internet and smart contracts, the size of Ethereum's state is also continuously growing. The larger the state, the higher the cost of node synchronization and maintenance. In addition, the growth of the state may also affect the cost of accessing the state in the future, as a large amount of data may lead to an increase in the time it takes to retrieve and process the data. Therefore, a conservative setting for Gas limits is also in place to control the speed of state growth, ensuring the long-term sustainability and performance of the network.

Monad Optimization

Monad has introduced optimizations in four major areas, enabling blockchain to achieve a throughput of 10,000 transactions per second (tps):

MonadBFT

onadBFT is a high-performance consensus mechanism used to achieve transaction ordering consensus under partially synchronous conditions in the presence of Byzantine participants. It is based on HotStuff and incorporates improvements from Jolteon/DiemBFT/Fast-HotStuff, particularly in utilizing secondary communication complexity in the case of leader timeouts to reduce three-round communication to two rounds.

Monad allows consensus and execution to be decoupled, enabling them to proceed independently. Delayed execution means that the execution of transactions can take place some time after they are included in the consensus process. This provides more flexibility for validating nodes, as they can execute transactions at a more appropriate time, rather than having to execute them immediately at consensus time.

Parallel Execution

Monad parallelizes transaction execution. Although this may seem to imply different execution semantics from Ethereum, it is not the case in reality. Monad blocks are the same as Ethereum blocks - both are linearly ordered collections of transactions. In both Monad and Ethereum, the results of executing transactions in a block are the same. Monad employs Optimistic Execution. This means that Monad will start executing subsequent transactions in a block even if previous transactions have not completed. Sometimes (but not always), this can lead to incorrect execution results. Although Monad parallelizes transaction execution, the state updates from each transaction are "merged" in order to prevent double-spending issues.

MonadDb

This is a custom database designed for storing blockchain state. It implements the MPT data structure natively and utilizes the latest asynchronous I/O technology and other optimization techniques to improve efficiency and performance.

Parallel Execution

Monad provides 1-second block time and 1-second finality. To achieve this, Monad has made some key changes to the consensus layer and execution layer (the two main mechanisms of blockchain), while preserving seamless compatibility for application developers (full EVM bytecode compatibility) and users (Ethereum RPC API compatibility).

Monad Labs co-founder and COO Eunice Giarta said: "Monad has redesigned the EVM execution system from scratch to allow non-overlapping transactions to run in parallel." She pointed out that the team has also reshaped the integration system between the two mechanisms to allow the execution layer and consensus of the blockchain to run in parallel.

Monad Lab co-founder and CEO Keone Hon stated in a statement: "Monad Blockchain will set industry standards for the speed, transparency, security, and scalability that all Layer 1 protocols should have."