The AWS Deep Learning Containers are available today with the latest framework versions of 1.15.2 with python 3.7 support . The release includes updates to the Amazon SageMaker Experiments package. Amazon SageMaker Experiments is a feature in Amazon SageMaker that lets you organize, track, compare, and evaluate machine learning (ML) experiments and model versions. The TensorFlow 1.15.2 python3.7 training containers now also include SageMaker Debugger , which allow data scientists to save and inspect the model tensors during training jobs.

The AWS Deep Learning Containers for Elastic Inference are available today with the framework versions PyTorch 1.3.1, TensorFlow 1.15.0, and TensorFlow 2.0.0. The PyTorch 1.3.1 upgrade includes the newly added SageMaker Inference and SageMaker PyTorch Inference. The TensorFlow 1.15.0 and TensorFlow 2.0.0 upgrades include the latest versions of TensorFlow Model Server for use with Elastic Inference. You can launch the new versions of the Deep Learning Containers on Amazon SageMaker, on Amazon EC2, and on Amazon Elastic Container Service (Amazon ECS). For a complete list of packages and versions supported by these Deep Learning Containers, see the release notes .

Today, Amazon WorkMail announced that you can now control email flow of your organization using AWS Lambda functions when using Email Flow Rules. With this, you can build powerful email flow control system with completely customizable conditions. For example, you can easily create Lambda to block any specific type of inbound or outbound email, or you can add or remove recipients to all or some inbound or outbound email. 

Amazon Kendra is now generally available to all AWS customers, with exciting new feature additions. Amazon Kendra provides customers with a highly accurate and easy to use enterprise search service powered by machine learning. Kendra offers a more intuitive way to search, using natural language, and returns more accurate answers; so your end users can discover information stored within the vast amount of content spread across your organization. Users can ask questions like “How long is maternity leave?” and get a specific answer such as “14 weeks”, or “How do I configure my VPN?” and get a specific passage extracted from a document describing the process. With Kendra, you can provide pinpoint search accuracy from content within your manuals, research reports, FAQs, HR documentation, customer service guides, and more. 

The AWS Deep Learning Containers are available today with the latest framework versions of PyTorch 1.5.0, with newly added SageMaker Inference, SageMaker PyTorch Inference, and the latest version of SageMaker PyTorch Training. You can launch the new versions of the Deep Learning Containers on Amazon SageMaker, Amazon Elastic Kubernetes Service (Amazon EKS), self-managed Kubernetes on Amazon EC2, and Amazon Elastic Container Service (Amazon ECS). For a complete list of frameworks and versions supported by the AWS Deep Learning Containers, see the release notes for PyTorch 1.5.0.

AWS Direct Connect makes it easy to establish a dedicated network connection from your premises to AWS. Using AWS Direct Connect, you can establish private connectivity between AWS and your data center, office, or colocation environment, which in many cases can reduce your network costs, increase bandwidth throughput, and provide a more consistent network experience than Internet-based connections.

Patch Manager, a capability of AWS Systems Manager, now allows you to deploy patches automatically to instances running Debian 8 (Jessie), Debian 9 (Stretch), and Oracle Linux 7.6, giving you more patching options for your mixed Linux environments. 

The Amazon Corretto team introduces Heapothesys, an open-source benchmark which simulates fundamental application characteristics that affect JVM GC latency. Heapothesys creates scenarios with pre-determined object allocation rates, heap occupancy, and heap sizes, then reports the resulting JVM pauses. The intent is to help OpenJDK developers investigate capability boundaries of the technologies they are implementing. It provides reference points for how different collector implementations perform when these basic stress factors are dialed up and collector leeway to act shrinks. We are working on enhancing Heapothesys to better model and predict additional application behaviors (see issue-12 ).

Starting today, Amazon EC2 M6g instances powered by Arm-based AWS Graviton2 processors are generally available. Amazon EC2 M6g instances deliver up to 40% better price performance over the current generation x86-based Amazon EC2 M5 instances for a broad set of general-purpose workloads including, application servers, microservices, gaming servers, small and mid-size databases, and caching fleets.

Earlier this year, we launched support for Aurora PostgreSQL user authentication with Kerberos and Microsoft Active Directory. In the original release, this support was based on AWS Directory Service for Microsoft Active Directory . We have now added support for user authentication using external Kerberos and Microsoft Active Directories, including those running on premises.