Analyzing Performance Tradeoffs in Sharding Blockchain System <Abstract> Blockchain faces scalability challenges that create bottlenecks in many applications. Sharding technology has emerged as a promising solution to enhance the scalability of blockchain systems. However, as the number of shards increases, the proportion of cross-shard transactions grows, becoming a critical bottleneck that limits overall system performance. Therefore, reducing the occurrence of cross-shard transactions is crucial. Current popular mechanisms for handling cross-shard transactions include Relay in Monoxide and the Broker role introduced in BrokerChain. Compared to Relay, Broker act as a third-party entity within shards to convert cross-shard transactions into intra-shard transactions, demonstrating superior performance. However, despite the benefits of introducing Broker in reducing cross-shard transaction ratios, research regarding the selection of number of brokers and its interaction with other key parameters (e.g., the number of shards and the number of nodes per shard) remains unexplored. These factors are vital in understanding their impact on system performance. Therefore, this study focuses on investigating the influence of key parameters (i.e., BrokerNum, ShardNum, and NodesInShard) on the performance metrics of sharding blockchain systems, in terms of Transactions Per Second (TPS), Transaction Confirmation Latency (TCL), and Cross-shard Transaction Ratio (CTX). Using the BlockEmulator platform, we conducted comprehensive experiments with different datasets. The results reveal that the performance of the sharding-based blockchain system exhibits trade-offs in the presence of these parameters. To achieve optimal system performance, further research and refinement in parameter configuration are essential. |