The key battleground for automotive stakeholders over the next three years will be achieving the production of electric, connected and autonomous vehicles at scale. Learning how to keep graphics processing units (GPUs) fueled with data when training the next generation of deep learning architectures is critical.
As GPU technology continues to advance, the demand for faster data continues to grow. In this 60-minute webinar, Ramnath Sai Sagar, AI and DL Product Lead at Pure Storage, presents a new benchmark suite for evaluating and tuning input pipelines. He examines results with TensorFlow’s DataSets API on a DGX-1 ‘supercomputer’ with V100 GPUs, and provide guidance on key tuning parameters and diagnostic techniques for improving performance.