AWS Deep Learning AMIs now with optimized TensorFlow 1.7 for faster training on Amazon EC2 C5 and P3 instances

The AWS Deep Learning AMIs for Ubuntu and Amazon Linux now come with TensorFlow 1.7, which is built with advanced optimizations for high-performance training across Amazon EC2 instance families. This is an update to the optimized build of TensorFlow 1.6 that we launched in late March.

Faster training with optimized TensorFlow 1.7

The Amazon Machine Images (AMIs) now come with TensorFlow 1.7 built with Intel’s Advanced Vector Instructions (AVX), SSE, and FMA instruction sets. The AMIs are fully configured with Intel’s Math Kernel Library for Deep Neural Networks (Intel MKL-DNN) to speed up training performance on Intel Xeon Platinum processors powering Amazon EC2 C5 instances. Training a ResNet-50 benchmark with a synthetic ImageNet dataset using our optimized build of TensorFlow 1.7 on a C5.18xlarge instance type was 9.8X faster than training on the stock TensorFlow 1.7 binaries.

The AMIs also come with an optimized build of TensorFlow 1.7 fully configured with NVIDIA CUDA 9 and cuDNN 7 to take advantage of mixed precision training on the Volta V100 GPUs powering Amazon EC2 P3 instances. A Deep Learning AMI automatically deploys the high performance build of TensorFlow optimized for the EC2 instance of your choice when you activate the TensorFlow virtual environment for the first time.

The AMIs also include TensorBoard 1.7 to help you visualize and debug your model training, and TensorFlow Serving 1.6 to quickly prototype an inference endpoint for your trained models.

Latest in deep learning frameworks

The Deep Learning AMIs now come with the Microsoft Cognitive Toolkit 2.5 with performance improvements and bug fixes. The AMIs also come with the latest in deep learning frameworks.

Frameworks with CUDA 9:

• Apache MXNet 1.1 (with Gluon)
• Caffe2 0.8.1
• Microsoft Cognitive Toolkit (CNTK) 2.5
• PyTorch 0.3.1
• TensorFlow 1.7
• Theano 1.0
• Chainer 3.5
• Caffe 1.0 with CUDA 8
• Keras 1.2.2 and Keras 2.1.5

The AMIs include model serving and debugging capabilities, provided by the following tools:

• Apache MXNet Model Server 0.1
• TensorFlow Serving 1.6
• TensorBoard 1.7

Getting started with the Deep Learning AMIs

It’s fast and simple to get started with the AWS Deep Learning AMIs. Our latest AMIs are now available on the AWS Marketplace. You can also subscribe to our discussion forum to get new launch announcements and post your questions.

About the Author

Sumit Thakur is a Senior Product Manager for AWS Deep Learning. He works on products that make it easy for customers to get started with deep learning on cloud, with a specific focus on making it easy to use engines on Deep Learning AMI. In his spare time, he likes connecting with nature and watching sci-fi TV series.