Search for: All records

The ever-growing demand for accurate machine learning models resulted in an increase in dataset and model sizes of deep neural networks. This paper discusses reconfigurable optical networks as the key enabler for scaling AI systems.

 

We present experimental demonstrations of ultra-low power edge computing enabled by wavelength division multiplexed optical links and time-integrating optical receivers. Initial experimentation demonstrations show ≲ 10 fJ of optical energy per MAC.

 

The bandwidth and latency requirements of modern datacenter applications have led researchers to propose various topology designs using static, dynamic demand-oblivious (rotor), and/or dynamic demand-aware switches. However, given the diverse nature of datacenter traffic, there is little consensus about how these designs would fare against each other. In this work, we analyze the throughput of existing topology designs under different traffic patterns and study their unique advantages and potential costs in terms of bandwidth and latency ''tax''. To overcome the identified inefficiencies, we propose Cerberus, a unified, two-layer leaf-spine optical datacenter design with three topology types. Cerberus systematically matches different traffic patterns with their most suitable topology type: e.g., latency-sensitive flows are transmitted via a static topology, all-to-all traffic via a rotor topology, and elephant flows via a demand-aware topology. We show analytically and in simulations that Cerberus can improve throughput significantly compared to alternative approaches and operate datacenters at higher loads while being throughput-proportional.