Yilong Li is a co-founder of MegaETH, a project focused on developing high-performance scaling solutions for blockchain networks. ^[7]

Overview

Yilong Li is involved in the development of MegaETH, a project aiming to address scalability challenges in blockchain ecosystems, particularly within the Ethereum Layer 2 context ^[1]. His background includes academic research in Computer Science at Stanford University ^[1].

MegaETH is being built with the goal of achieving high transaction throughput by leveraging underlying blockchain layers like Ethereum and data availability layers such as EigenDA. The project utilizes specific technical approaches, including advancements in authenticated key-value stores, to enhance performance. ^{[2] [3]}

Background

Prior to his work on MegaETH, Yilong Li was a PhD student in Computer Science at Stanford University. ^[7]

MegaETH

MegaETH is a project co-founded by Yilong Li alongside Shuyao Kong. The project focuses on creating high-performance scaling solutions, specifically targeting Ethereum Layer 2 networks ^[2].

Key aspects of MegaETH include:

• Scaling Approach: The project's technologies are designed for scaling Layer 2 solutions, leveraging the security and censorship resistance provided by base layers like Ethereum ^[2].
• Performance Targets: MegaETH aims to scale an EVM-based chain to over 100,000 transactions per second (TPS) ^[2]. With the integration of technologies like Quick Merkle Database (QMDB), the project has discussed the potential for 1 million TPS, benchmarked using workloads like ERC-20 transfers ^[4] .
• Data Availability: MegaETH utilizes EigenDA for data availability, citing its hyperscale throughput capabilities as a key factor in this choice ^[3]. The selection is based on EigenDA being considered a near-optimal data availability design for Ethereum Layer 2s from an information-theoretic perspective ^[3].
• Core Technology: The project incorporates advancements in authenticated key-value stores, such as Quick Merkle Database (QMDB) . QMDB is described as a high-performance verifiable database optimized for blockchains, capable of high state update rates and transaction throughput ^[6].
• Architecture: From the perspective of a replica node, MegaETH involves receiving state differences (state diffs) from a peer-to-peer (P2P) network and relying on external proofs for state transitions ^[4].
• Development Status: A public testnet for MegaETH was announced with a target launch within 30 days from February 3, 2025 ^[6].

Views and Interactions

Yilong Li has publicly discussed various technical topics related to blockchain scaling and performance.

• On Layer 2 vs. Layer 1 Scaling: He has stated that many of the scaling technologies developed for MegaETH, which is focused on Layer 2, do not directly apply to Layer 1 blockchains like Ethereum. This is because Layer 2 solutions can leverage the security and censorship resistance of the base layer, enabling more aggressive optimizations ^[2].
• On Quick Merkle Database (QMDB): Li has expressed interest in QMDB, describing it as a significant breakthrough in authenticated key-value stores with a simple and elegant design. He reviewed a draft paper on QMDB and congratulated the LayerZero team on its development. ^[1]
• On Data Availability Choices: He explained that EigenDA was chosen for MegaETH due to its capacity for hyperscale throughput, which is considered necessary to realize the vision of Ethereum as a "world computer" ^[3].