Key points are not available for this paper at this time.
Abstract This study systematically investigates the performance scalability of Raspberry Pi 4 Model B clusters across various configura-tions. Leveraging the latest advancements in the Pi4B architecture, including improved processor performance, increased memory capacity, and enhanced network bandwidth, we explore Raspberry Pi clusters’ computational efficiency and cost-effectiveness. We use the High-Performance Linpack (HPL) benchmark to measure the clusters’ floating-point operations per second (FLOPS) across diverse node combinations and memory usage scenarios, examining cluster scalability andperformance thresholds. Our findings reveal a nuanced, nonlinear scaling behavior of performance with increasing cluster sizes. This paper contributes valuable insights for designing and optimizing Raspberry Pi clusters in scalable computing applications,offering a foundational perspective for future exploration in this rapidly evolving field.
Liu et al. (Wed,) studied this question.