machine-learning video for a healthcare analytics company is using Amazon SageMaker to train a deep learning model for medical image analysis. The training
A healthcare analytics company is using Amazon SageMaker to train a deep learning model for medical image analysis. The training process requires distributed training across multiple GPU instances due to the large size of the dataset. During training, the ML engineer observes that the model training is slower than expected because of network communication delays between the instances. The engineer must optimize the network configuration to minimize communication overhead and improve training performance. Which of the following options should be combined for a solution that meets these requirements? (Select two)