Abstract:

With the widespread application of Electroencephalogram (EEG) signal processing technology in medical, neuroscience, and human-computer interaction fields, traditional single-machine architectures face challenges such as limited computational resources, high latency, and poor stability in high-concurrency scenarios. This paper proposes a high-performance EEG signal processing cluster architecture based on Nginx load balancing. By leveraging Nginx’s reverse proxy technology, concurrent requests are distributed to distributed nodes, combined with standardized data processing workflows and a unified cluster entry point, significantly improving real-time performance of system, resource utilization, and throughput. Experimental results demonstrate that under 100~2 000 concurrent loads, the cluster architecture achieves notable optimizations in both average and minimum response time, along with a substantial increase in throughput. Additionally, Nginx’s IP access control strategy effectively blocks unauthorized requests, ensuring data security. The architecture enables efficient resource scheduling through dynamic load balancing, providing robust technical support for large-scale neural data processing. Future work will integrate artificial intelligence and edge computing technologies to extend the architecture to complex EEG analysis scenarios, advancing precision medicine and intelligent interaction development.

Key words: Nginx, load balancing, cluster, EEG signal processing, secure access