FOSquare: A Novel Optical HPC Interconnect Network Architecture Based on Fast Optical Switches with Distributed Optical Flow Control

Interconnecting networks adopting Fast Optical Switches (FOS) can achieve high bandwidth, low latency, and power consumption. We propose demonstrate a novel interconnecting topology based on FOS (FOSquare) with distributed fast flow control which is suitable for HPC infrastructures. also present an Optimized Mapping (OPM) algorithm that maps the most communication-related processes inside rack. numerically investigate compare network performance of FOSquare Leaf-Spine under real traffic traces collected by running multiple applications (CG, MG, MILC, MINI_MD) in infrastructure. The numerical results show reduce >10% latency respect to scenario 16 available cores.