Note: This video may require joining the NVIDIA Developer Program or login
GTC Silicon Valley-2019 ID:S9212:Distributed Deep Learning
One of the key advantages of the cloud is that it allows users to scale out compute resources as required. We'll examine distributed training on deep learning models, which is tricky because it requires understanding the frameworks used and the computing infrastructure. We'll discuss benchmarking carried out on the Azure platform on different GPU architectures and across a number of frameworks.