ISE Blog

AWS re:INVENT 2017 Announcements & Enhancements

It’s that time of the year again when the cloudy masses descend on Las Vegas, Nevada for another Amazon Web Services (AWS) re:Invent Conference. With over 1,800 breakout sessions and more than 46,000 attendees from all over the globe, the sixth annual conference is the biggest yet, spanning five venues across the Vegas Strip. This year, I’m writing this post live from the conference to preview some insights from the largest gathering of cloud experts, engineers and advocates in the world, as they unfold.

AWS Conference 2017
The key theme of this year’s new product announcements is extending serverless offerings with serviceful features. See the highlights below to see an overview of how AWS is bringing more great features to developers as services.

Product Announcements

AWS’s renewed commitment to video and other media showed strongly in the new products they announced Monday, with five new elemental media services, and a new Augmented Reality and Virtual Reality (AR & VR) offering: AWS Sumerian. Sumerian is a tool enabling rapid development of AR and VR content that works across different types of common AR and VR devices. You can create 3D environments and experiences right in your browser.

Elemental Media Services

The Elemental services announced span the needs of a video delivery system. Taken together these services allow customers to build complete, monetized, on-demand video applications that can be viewed on any device anywhere.

Tuesday Night Live

Peter Desantis took time during his Tuesday night keynote to dive deep into the evolution of the new hardware underpinning the EC2 service. The introduction of M5 instances represent the culmination of a several-year effort to bring the EC2 platform services into a hardware layer in the datacenter, meaning better performance, lower variability and greater overall customer capacity for EC2. The new generation of AWS EC2 instances run on the custom ASIC hardware designed for AWS specifically for the special needs of AWS and their customers by Annapurna Labs. The result is a new platform that allows AWS to expose "bare metal" instances to customers for use with restrictive 3rd party licensed products, specific virtualization kernels, and more.

AWS Conference 2017
Keynote Announcements

AWS CEO, Andy Jassy’s keynote, one of the highlights of the event, rolled out a ton of new features with a real emphasis on enabling services for developers to integrate the newest technologies into their applications and workloads without having to provision and manage servers. Jassy announced many new and preview features.

Elastic Container Service for Kubernetes and Fargate

Finally, the Elastic Container Service has been extended to encompass a fully managed Kubernetes installation. Kubernetes is an important service deployment framework with containers serving as the runtime. The managed Kubernetes service allows you to deploy a Kubernetes cluster easily across multiple availability zones for no single point of failure. And whether you deploy your containers on Kubernetes or not, you can now use the Fargate service to deploy your containers without managing servers.

AuroraDB Enhancements

Aurora is one of the fastest growing services in AWS history, and serves as a great high capacity, high availability relational database service. The addition of Aurora Master Replicas allows simultaneous scaling of read and write, and provides zero downtime failover for high-availability Aurora deployments. Continuing the theme of serviceful offerings, the preview of Aurora Serverless allows all of the great features of Aurora in a pay-as-you-go managed service.


One of AWS’s earliest offerings, the NoSQL database Dynamo now offers global-replication for tables to be available to your application with great performance wherever in the world you might be accessing it. Dynamo backup and restore adds critical features with zero-downtime snapshots and, coming soon, will add point-in-time restores.

Machine learning

Some of the most advanced AWS features rolled out are in the machine learning (ML) space. Sagemaker aims to simplify the entire pipeline of training and deploying a ML model, giving modular flexibility to deploy models trained elsewhere or to train on AWS and deploy on premises. This middle tier of ML services to enable easier custom development should spawn a new era in machine learning applications.

At the top tier, providing easily-integrated ML services to developers with minimal fuss, Rekognition extends AWS’ image recognition services to video, and integrates with a new type of Kinesis stream specifically for ingesting video from disparate devices and sources. AWS Transcribe and AWS Translate add full text recognition and translation to the video analysis services. And Amazon Comprehend provides a fully-managed natural language processing (NLP) service for entity recognition and textual analysis.

AWS Conference 2017
IoT Features

Rounding out the new features announced, are enhancements to AWS IoT capabilities, enabling management of large fleets of IoT devices with IoT device manager, and security enhancements through IoT device defender.  In addition, there's a new IoT analytics offering!  The IoT advancements are enough to merit their own future post.  Stay tuned!

Still More to Come

Doubtlessly, Werner Vogels will announce even more tremendous features at tomorrow’s keynote. If you aren’t at the conference, you can watch the keynote live on AWS’s reinvent live stream. It’s great to see the advancements the cloud market leader is bringing for every developer to be able to add unprecedented capabilities to their applications and workloads with greater ease and flexibility than ever before.

Thoughts on the latest annoucements from the conference? Comment below and let us know what you think! Still can't get enough AWS? Check out last year's conference recap.

Samuel Thurston, Software Engineer

Samuel Thurston, Software Engineer

Samuel Thurston is a Software Engineer and Cloud Practice Lead for ISE, architecting and implementing cloud solutions for enterprise clients. He enjoys running, yoga, and cooking, and is frequently found on the disc golf course.

Samuel Thurston, Software Engineer

Latest posts by Samuel Thurston, Software Engineer (see all)

What’s Wrong with Monoliths? Oct 25, 2018

FaaS: Function as a Service Sep 28, 2018

Networking is Hard Aug 23, 2018

Observability in Distributed Systems Jul 19, 2018

Hands-On With Amazon AWS DeepLens Jun 21, 2018