WHAT'S NEW AT AWS re:Invent 2017

Returning from yet another week packed with experiences from AWS re:Invent, I can't help but report back that AWS was ~90% about machine learning or artificial intelligence, and it’s clear that AWS is moving up the stack.

Here's are some of the announcements highlights:

AWS FARGATE

AWS Fargate is a fully managed container service that allows you to run containers on EC2 without having to manage instances or clusters. This is welcome news to customers have been responsible for managing the underlying capacity, availability, etc.

ELASTIC CONTAINER SERVICE FOR KUBERNETES

Many of our customers run Kubernetes on AWS. But this has never been the easiest process, and has always required a lot of manual configuration and resulted in some difficult challenges. That changes with the introduction of Elastic Container Service for Kubernetes (EKS for short).

EKS allows you to run a managed Kubernetes master tier within your AWS cloud. Amazon EKS automatically runs Kubernetes with three masters across three AZs, and automatically detects and replaces unhealthy masters. Kubernetes already integrates natively with Elastic Load Balancing, Security Groups, and VPCs, so that your engineers no longer have to custom engineer a solution.

MULTI-MASTER AURORA AND AMAZON AURORA SERVERLESS

Although this announcement didn’t make the big headlines, it’s very big news for anyone running Aurora. Amazon Aurora Serverless in an auto-scaling configuration for Aurora that allows you to run your database in the cloud without managing any database instances. It’s perfect for workloads that have intermittent usage, as it can start up, shut down, and scale up or down based on your application.

AMAZON SAGEMAKER

Perhaps the most widely-discussed announcement out of Re:Invent, Amazon SageMaker is yet another service that packages together all of AWS’ previously-announced machine learning capabilities into an easy-to-use, fully-managed service.

SageMaker is designed to empower any developer to use machine learning, making it easy to build and train models and deploy them to production. It’s one-click machine learning for the masses.

WELL-ARCHITCTED PRINCIPLES

AWS is heavily encouraging its customers to conduct quarterly or bi-annual Well-Architected Reviews. It’s a formal process that is performed by AWS Professional Services and a limited number of approved Well-Architected Framework delivery partners.