Home > AI Solutions > Artificial Intelligence > White Papers > Model Customization for Code Creation with Red Hat OpenShift AI on Dell AI Optimized Infrastructure > Hardware design
Although the inferencing and training activities described in this document use only a single node at any point in time, for multi-node AI activities, networking plays a critical role in interconnecting GPU servers to optimize the GPU fabric that enhances the ability of the environment to perform the distributed AI tasks. In addition, for all types of AI environments, loading AI models from storage and keeping temporary checkpoints in storage using frontend fabric during model training means that network component selection is a critical element of any architectural design. Dell’s flagship PowerSwitch products for GenAI solutions are the Z9664F and Z9864F.