Timely real-time processing of electrocardiogram (ECG) signals imposes a significant computational burden due to the complexity and volume of biomedical datasets. This study proposes an OpenMP-accelerated parallel and distributed ECG processing framework aimed at optimizing performance for real-time large-scale ECG analysis. We implemented and parallelized three core algorithms - Moving Average Filtering, Naive Bandpass Filtering, and Dynamic Peak Detection - across multiple threads using OpenMP. A case study on thread scaling was conducted on Google’s v5e−1 TPU, analyzing execution from 1 to 32 threads. The most notable results were obtained with 6 threads, achieving a 5.3× speedup and reducing sequential execution time from 62,908 ms to 11,962 ms. Beyond 8 threads, performance improvements stagnate, indicating a saturation point. Further enhancements were explored through compiler optimizations (-O3, -fopenmp-simd, -march = native, and -funroll-loops) and alternative thread scheduling strategies (dynamic, static, and guided). Overall, the proposed framework effectively mitigates computational costs and establishes a scalable, optimized, and accessible ECG monitoring system for real-time edge computing environments.