AERO: Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration

Mingxuan Song, Pengze Li, Bohan Zhou, Shenglin Yin, Zhen Xiao, Jieyi Long

Track: Systems and infrastructure for Web, mobile, and WoT

Keywords: Blockchain, Sharding, Account migration, Reinforcement learning

TL;DR: This paper proposes AERO, a deep reinforcement learning framework for efficient account migration in sharding blockchains, improving throughput by 31.77% and reducing cross-shard transactions and workload imbalances.

Abstract: Sharding blockchain networks face significant scalability challenges due to high frequencies of cross-shard transactions and uneven workload distributions among shards. To address these scalability issues, account migration offers a promising solution. However, existing migration solutions struggle with the high computational overhead and insufficient capture of complex transaction patterns. We propose AERO, a deep reinforcement learning framework for efficient account migration in sharding blockchains. AERO employs a prefix-based grouping strategy to enable group-level migration decisions and capture complex transaction patterns and relationships between accounts. We also implement a sharding blockchain system called AEROChain, which integrates our decentralized AERO and aligns with the blockchain decentralization principle. Extensive evaluation with real Ethereum transaction data demonstrates that AERO improves the system throughput by 31.77% compared to existing solutions, effectively reducing cross-shard transactions and balancing shard workloads.

Submission Number: 844