Blockchain performance is crucial in research, with sharding emerging as an effective solution for scalability. By dividing the network into smaller shards, sharding facilitates faster transaction processing. However, there is currently a lack of effective simulators for modeling sharding blockchain performance and assessing shard load balancing. In this paper, we introduce ShardingSim, a modular, committee-based sharding blockchain simulator. ShardingSim simulates various sharding configurations and evaluates their performance across diverse network conditions and transaction datasets. We present a use case by modeling and simulating RapidChain with ShardingSim: simulation results show that RapidChain’s performance improves with more shards under historical Bitcoin transaction datasets; however, such enhancement is absent in scenarios with uneven transaction distributions, highlighting the need for additional load balancing methods. Through ShardingSim, we can simulate committee-based sharding blockchain, evaluateits performance, and identify potential load-balancing issues between shards.