There is essentially no absolute leader in the competition among Monad, MegaETH, and Pharos. Developers are left with more trade-offs to consider in terms of performance, decentralization, or specialization priorities.

Introduction

Recently, the three major parallel EVM projects have launched their testnets almost simultaneously: Monad launched its testnet on February 19, MegaETH on March 21, and Pharos on March 24. The main narrative of Web3 technology seems to have returned to parallel EVM after AI Agent, which was the hottest narrative at the beginning of 2024.

The Parallel EVM Holy Grail Contention: Monad, MegaETH, and Pharos

The Ethereum Virtual Machine (EVM) is the core of Ethereum, responsible for running smart contracts and processing transactions. The EVM is a computing engine that provides abstractions for computation and storage. However, the EVM lacks scheduling capabilities. Ethereum's execution module takes transactions from blocks one by one, and the EVM executes them sequentially. Although sequential execution ensures that transactions and smart contracts can be executed in a deterministic order, ensuring security, it can lead to network congestion and latency under high load.

Parallel EVMs significantly enhance the throughput of the network by allowing multiple operations to execute simultaneously, thereby improving the performance and scalability of the entire blockchain. In fact, what we refer to as parallel EVMs are high-performance EVM-compatible blockchains that not only introduce parallel execution but also feature comprehensive upgrades from consensus, transactions, pipelining, storage to hardware acceleration. The goal is to enable blockchain networks to process more transactions in a shorter amount of time, effectively solving the traditional blockchain issues of network congestion and latency.

This article will delve into the backgrounds and architectures of Monad, MegaETH, and Pharos, as well as the trade-offs developers face in choosing among them.

Monad

Monad is a high-performance EVM-compatible Layer 1 blockchain developed by Monad Labs. Monad aims to improve system scalability while maintaining decentralization, addressing the low throughput issues of existing EVM-compatible blockchains.

Monad Labs was co-founded in 2022 by Keone Hon, James Hunsaker, and Eunice Giarta, with Keone and James being former employees of the market-making giant Jump Trading, and Eunice coming from a non-Crypto background.

In February 2023, Monad Labs completed a $19 million seed round led by Dragonfly. In April 2024, Monad Labs secured another $225 million in funding led by Paradigm. Monad's current valuation has reached $3 billion.

Monad's key strengths lie in its ability to handle up to 10,000 transactions per second (TPS) with a block time of just 1 second. This is primarily due to optimizations in the following four areas:

MonadBFT: A high-performance consensus mechanism, an improved version of HotStuff. It is used to reach consensus on transaction ordering under partial synchrony in the presence of Byzantine actors. First, MonadBFT employs a two-stage BFT algorithm with optimistic responsiveness and linear communication overhead in common cases, and quadratic communication overhead in case of timeouts. Second, MonadBFT uses a hybrid signature scheme, with the integrity and authenticity of messages provided by ECDSA signatures, and aggregatable message types (votes and timeouts) provided by BLS signatures, addressing scalability issues. Moreover, in Monad, nodes do not maintain a global transaction pool but a local one, with transactions forwarded by RPC nodes to subsequent leader nodes, effectively reducing bandwidth usage and transaction latency. Lastly, MonadBFT's message propagation utilizes the RaptorCast protocol, which not only converts block proposals into erasure-coded blocks but also sends each block to all validators via a two-tier broadcast tree. RaptorCast leverages the entire network's upload bandwidth to propagate block proposals to all validators while retaining Byzantine fault tolerance. Thanks to these features, MonadBFT achieves efficient and robust blockchain consensus.

Asynchronous Execution: Asynchronous execution allows Monad to significantly increase execution throughput by decoupling consensus and execution. By separating consensus and execution, Monad can substantially increase the execution budget, as execution shifts from occupying a small fraction of block time to occupying the entire block time. First, Monad block proposals do not include state roots. To prevent node forking, block proposals also include the state root from three blocks prior, allowing nodes to detect if they have forked. Second, leader nodes construct blocks with a delayed state perspective. To defend against DDoS attacks, Monad nodes verify that the account balance is sufficient to cover the highest debit in the user's account associated with the ongoing transaction. Lastly, upon receiving the proposed block, although the block is not yet finalized, nodes can still locally execute the proposed block during this period, but there is no guarantee that it will be voted on or finalized. Thanks to these features, Monad achieves significant speed improvements, enabling a single-shard blockchain to scale to millions of users.

Parallel Execution: Monad employs an optimistic execution approach, starting the execution of subsequent transactions before the completion of earlier transactions in a block, with each transaction's updated state merged in sequence. This can sometimes lead to incorrect execution results. To address this, Monad tracks the inputs used during transaction execution and compares them with the outputs of previous transactions. If there are discrepancies, it indicates that the transaction needs to be re-executed with the correct data. Moreover, Monad uses a static code analyzer during transaction execution to predict transaction dependencies

There is essentially no absolute leader in the competition among Monad, MegaETH, and Pharos. Developers are left with more trade-offs to consider in terms of performance, decentralization, or specialization priorities.

Introduction

Recently, the three major parallel EVM projects have launched their testnets almost simultaneously: Monad launched its testnet on February 19, MegaETH on March 21, and Pharos on March 24. The main narrative of Web3 technology seems to have returned to parallel EVM after AI Agent, which was the hottest narrative at the beginning of 2024.

The Parallel EVM Holy Grail Contention: Monad, MegaETH, and Pharos

The Ethereum Virtual Machine (EVM) is the core of Ethereum, responsible for running smart contracts and processing transactions. The EVM is a computing engine that provides abstractions for computation and storage. However, the EVM lacks scheduling capabilities. Ethereum's execution module takes transactions from blocks one by one, and the EVM executes them sequentially. Although sequential execution ensures that transactions and smart contracts can be executed in a deterministic order, ensuring security, it can lead to network congestion and latency under high load.

Parallel EVMs significantly enhance the throughput of the network by allowing multiple operations to execute simultaneously, thereby improving the performance and scalability of the entire blockchain. In fact, what we refer to as parallel EVMs are high-performance EVM-compatible blockchains that not only introduce parallel execution but also feature comprehensive upgrades from consensus, transactions, pipelining, storage to hardware acceleration. The goal is to enable blockchain networks to process more transactions in a shorter amount of time, effectively solving the traditional blockchain issues of network congestion and latency.

This article will delve into the backgrounds and architectures of Monad, MegaETH, and Pharos, as well as the trade-offs developers face in choosing among them.

Monad

Monad is a high-performance EVM-compatible Layer 1 blockchain developed by Monad Labs. Monad aims to improve system scalability while maintaining decentralization, addressing the low throughput issues of existing EVM-compatible blockchains.

Monad Labs was co-founded in 2022 by Keone Hon, James Hunsaker, and Eunice Giarta, with Keone and James being former employees of the market-making giant Jump Trading, and Eunice coming from a non-Crypto background.

In February 2023, Monad Labs completed a $19 million seed round led by Dragonfly. In April 2024, Monad Labs secured another $225 million in funding led by Paradigm. Monad's current valuation has reached $3 billion.

Monad's key strengths lie in its ability to handle up to 10,000 transactions per second (TPS) with a block time of just 1 second. This is primarily due to optimizations in the following four areas:

MonadBFT: A high-performance consensus mechanism, an improved version of HotStuff. It is used to reach consensus on transaction ordering under partial synchrony in the presence of Byzantine actors. First, MonadBFT employs a two-stage BFT algorithm with optimistic responsiveness and linear communication overhead in common cases, and quadratic communication overhead in case of timeouts. Second, MonadBFT uses a hybrid signature scheme, with the integrity and authenticity of messages provided by ECDSA signatures, and aggregatable message types (votes and timeouts) provided by BLS signatures, addressing scalability issues. Moreover, in Monad, nodes do not maintain a global transaction pool but a local one, with transactions forwarded by RPC nodes to subsequent leader nodes, effectively reducing bandwidth usage and transaction latency. Lastly, MonadBFT's message propagation utilizes the RaptorCast protocol, which not only converts block proposals into erasure-coded blocks but also sends each block to all validators via a two-tier broadcast tree. RaptorCast leverages the entire network's upload bandwidth to propagate block proposals to all validators while retaining Byzantine fault tolerance. Thanks to these features, MonadBFT achieves efficient and robust blockchain consensus.

Asynchronous Execution: Asynchronous execution allows Monad to significantly increase execution throughput by decoupling consensus and execution. By separating consensus and execution, Monad can substantially increase the execution budget, as execution shifts from occupying a small fraction of block time to occupying the entire block time. First, Monad block proposals do not include state roots. To prevent node forking, block proposals also include the state root from three blocks prior, allowing nodes to detect if they have forked. Second, leader nodes construct blocks with a delayed state perspective. To defend against DDoS attacks, Monad nodes verify that the account balance is sufficient to cover the highest debit in the user's account associated with the ongoing transaction. Lastly, upon receiving the proposed block, although the block is not yet finalized, nodes can still locally execute the proposed block during this period, but there is no guarantee that it will be voted on or finalized. Thanks to these features, Monad achieves significant speed improvements, enabling a single-shard blockchain to scale to millions of users.

Parallel Execution: Monad employs an optimistic execution approach, starting the execution of subsequent transactions before the completion of earlier transactions in a block, with each transaction's updated state merged in sequence. This can sometimes lead to incorrect execution results. To address this, Monad tracks the inputs used during transaction execution and compares them with the outputs of previous transactions. If there are discrepancies, it indicates that the transaction needs to be re-executed with the correct data. Moreover, Monad uses a static code analyzer during transaction execution to predict transaction dependencies