TransShard: A Dynamic Transaction-Aware Sharding Scheme for Account-Based Blockchain
Qian Wang, Yepeng Guan · IEEE Access · 2024
The poor scalability of blockchain technology restricts its application in large-scale networks. Sharding technology is viewed as the most promising on-chain solution to improving blockchain scalability. However, the high proportion of cross-shard transactions (TXs) and imbalanced transaction load on different shards resulting from the previous sharding schemes prevent further performance improvements. To address this problem, a dynamic transaction-aware sharding scheme (TransShard) is proposed for account-based blockchain. Firstly, a token-based cryptographic sortition approach is introduced to form shards. Nodes are weighted based on the tokens they own and assigned to different shards randomly in a private and non-interactive way. The design of this approach is the basis of transaction allocation and ensures the security of the whole blockchain. Secondly, an on-the-fly transaction placement algorithm is designed to optimize initial transaction allocation. The blockchain transactions are modeled as a dynamic graph, and a new scoring heuristic is introduced to find the optimal shards for incoming transactions and determine the placement of new accounts. Lastly, an incremental account reallocation algorithm is designed to maintain the quality of account allocation as the transaction graph updates. When the allocation quality is suboptimal, some of the hot accounts are examined to determine how to migrate their ledger states. Simulation results show that the proposed scheme has some excellent performances compared to some other state-of-the-art sharded blockchains in system throughput, transaction confirmation latency, and workload balance.