AWS’s annual re:Invent conference kicked off earlier this week in Las Vegas. The yearly event provides the cloud computing community with opportunities for learning through keynotes, training and certification opportunities, and technical sessions.

Here are a few of the announcements Amazon has made at the event so far: 

Amazon goes in on quantum computing
Quantum computing has gotten a lot of attention lately, especially since Google reached a major milestone with quantum computing back in October. At re:Invent, Amazon announced several capabilities for quantum computing, including Amazon Braket, AWS Center for Quantum Computing, and Amazon Quantum Solutions Lab. 

Amazon Braket is a managed server that will allow scientists, researchers, and developers to begin experimenting with multiple quantum computers from a single place. 

RELATED CONTENT: AWS re:Invent: Matillion Data Loader, Node.js functions for AWS Lambda, and new AWS support

AWS Center for Quantum Computing is a research center adjacent to Caltech that will bring together quantum computing researchers and engineers. Their goal with the center is to accelerate the development of quantum computing hardware and software.

Finally, Amazon Quantum Solutions Labs is a program that connects AWS customers with quantum computing experts from Amazon or a set of consultants. 

AWS end-of-support migration program for Windows Server
AWS has also announced a new program to help IT teams future-proof their end-of-support Windows Server Applications. Support for Windows Server 2008 and Windows Server 2008 R2 will end on January 14, 2020, meaning that applications running on those after that date will be vulnerable. 

The AWS End-of-Support Migration Program (EMP) for Windows Server program provides a combination of guidance and technology to help customers migrate from unsupported Windows servers to supported AWS ones. 

AWS introduces Bring-Your-Own-License experience for Microsoft Windows Server and SQL Server
The company also introduced an easier way to bring and manage existing Microsoft Windows Server and SQL Server licenses to AWS. Now, customers can use use pay-as-you-go licenses from AWS to launch Windows Server or SQL Server instances. 

According to the company, when customers obtain these licenses from AWS, they will also get a “fully-compliant, pay-as-you-go licensing model that doesn’t require managing complex licensing terms and conditions.” 

AWS releases Amazon SageMaker Operators for Kubernetes
AWS has announced the release of Amazon SageMaker Operators for Kubernetes. This new service will make it easy for data scientists to train, tune, and deploy machine learning models on Kubernetes in Amazon SageMaker.

Customers can now install Amazon SageMaker Operators on Kubernetes clusters. This will allow them to create SageMaker jobs using the Kubernetes API and command-line Kubernetes tools. 

There are currently three usable Amazon SageMaker Operators: Train, Tune, and Inference. These allow users to train their models in SageMakers, tune model hyperparameters, and deploy trained models to deliver high performance and available for prediction.  

AWS introduces other new features in SageMaker
In addition, AWS has introduced several new features for Amazon SageMaker, including SageMaker Debugger, SageMaker Model Monitor, SageMaker Processing, SageMaker Autopilot, and SageMaker Experiments. 

  • SageMaker Debugger helps with debugging machine learning models.
  • SageMaker Model Monitor provides fully managed monitoring of machine learning models in production, alerting customers when there is an issue.
  • Amazon SageMaker Processing provides fully managed data processing and model evaluation.
  • Amazon SageMaker Autopilot allows developers to create high-quality machine learning models and have full control and visibility over them.
  • Amazon SageMaker Experiments allows organizations to organize, track, compare, and evaluate machine learning experiments and models. 

AWS DeepComposer now available
Another machine learning capability the company introduced at re:Invent is AWS DeepComposer. AWS DeepComputer is a machine learning-enabled musical keyboard. 

The 32-key, 2-octave keyboard is designed to allow developers to play around with Generative AI. Developers can use either a pretrained model, or train their own model on a data set from their favorite genre. 

AWS Single Sign-On now integrates with Azure AD
AWS Single Sign-On has also gotten an update that allows Azure Active Directory (AD) customers to use their existing identity store with AWS Single Sign-On.

In addition, AWS added support for automatic synchronization of user identities and groups from Azure AD. Users will now be able to sign into accounts and applications in AWS environments with their existing Azure AD identity. 

Now, administrators will only have to manage a single source of truth for user identities, while also being able to configure access to all AWS accounts and apps centrally. 

Access Analyzer for S3 now available
The company also announced Access Analyzer for S3. Access Analyzer for S3 helps monitor access policies to ensure that they are providing intended access to S3 resources.

The service evaluates bucket access policies to help customers discover and remediate buckets with unintended access. 

AWS Outposts can now be ordered
Last year at re:Invent, AWS introduced AWS Outposts, and now the company is ready to take orders for them. Outposts are fully managed AWS infrastructure and services. It is intended for customers who need local processing and low latency. 

When a customer orders AWS Outposts, AWS technicians will come install, connect, set up, and verify them. After installation, AWS will continue to monitor, maintain, and upgrading them. Outposts also feature modular hardware, which allows hardware to be swapped out without downtime. 

Amazon Detective announced as preview feature
Finally, Amazon announced a preview of Amazon Detective, which is a service that helps customers identify the root causes of security issues. It automatically collects log data from AWS resources and then uses machine learning, statistical analysis, and graph theory to build a linked set of data that can be used in security investigations.