Multidomain Inter/Intrachip Silicon Photonic Networks for Energy-Efficient Rack-Scale Computing Systems

Abstract

Rack-scale computing systems are promising to undertake the emerging large-scale applications by distributing massive tasks to processing cores. The communication and coordination efficiency of these tasks and resources directly affect the system performance and energy consumption. Silicon photonic interconnects are expected to address the communication and system power consumption challenges imposed on rack-scale systems. However, the control for optical interconnects can cause server performance degradation if not properly designed, especially for the complicated and time-consuming multi-domain networks. In this paper, we study the optical interconnects for rack-scale computing systems and propose a new communication flow and control scheme for the efficient coordination of distributed resources. Particularly, we first propose a forward propagation strategy that parallels the path reservation process with the distributed tasks connection setup. Second, we develop a preemptive chain feedback (PCF) scheme to optimize multi-domain path reservation. PCF scheme preemptively allocates network resources with the help of multi-cell reservation window and quickly releases resources with a feedback mechanism. This solution increases network resources utilization and task coordination efficiency while minimizing path reservation overheads. Comparing to the baseline InfiniBand network fabric and handshake scheme, PCF can improve network throughput greatly under uniform and hotspot traffic patterns. Realistic benchmarks results show that PCF scheme on average reduces 52% and 60% energy consumption per unit system performance than InfiniBand and handshake scheme for a 256-node rack system.