Adaptive and efficient shard load balancing for high-throughput blockchain systems
摘要
Blockchain technology holds immense transformative potential but often faces significant scalability issues that impede its broader adoption. This study introduces a novel strategy to enhance blockchain scalability through optimized Load Balancing (LB), achieved by distributing nodes across shards using a customized M/M/k queuing model. The proposed method employs distinct distribution mechanisms for incoming and outgoing queries, ensuring balanced processing, reducing idle time, and improving network traffic management. Extensive simulations demonstrate a 25-32% increase in throughput, 18% reduction in response time, and 21% improvement in shard utilization compared to Round-Robin (RR), Least Connection (LC), Weighted RR (WRR), and Weighted LC (WLC). This solution offers a foundation for deploying scalable, reliable, and high-performance blockchain systems applicable to e-commerce platforms, gaming, credential verification, and cross-border payments.