ACE-AI™

AI workloads are increasingly distributed

Distributed AI offers significant computational efficiency, scalability, security, and latency benefits. Examples of how AI is being distributed across the network include Distributed Model Training, where AI/ML models are trained on multiple nodes within the network, enhancing efficiency and performance for large, complex models. Federated Learning is another approach, where AI/ML models are trained on data distributed across the network and multiple device types, including smartphones, tablets, and wearables. Additionally, Inferencing at the Edge involves deploying inferencing models at the edge of the network, closest to end users, which reduces latency and improves performance for applications. Key requirements for networks supporting distributed AI include high performance and lossless connectivity, predictable latency, high availability and resiliency with zero impact failover, and fabric-wide visibility.