At its annual conference, Amazon Web Service unveiled new services for Kubernetes, bare metal compute instances, and more.
On Tuesday, at the 2017 re:Invent conference in Las Vegas, Amazon Web Services (AWS) showed off a host of new products and services advancing its efforts in containers, databases services, and compute.
One of the first big announcements made by AWS CEO Andy Jassy was Amazon Elastic Container Service for Kubernetes (EKS), a managed Kubernetes integration that is hybrid cloud compatible and provides automatic updates and patches. EKS will exist as an alternative to Amazon Elastic Container Service (ECS), Amazon's initial attempt at container management.
For customers that want to run containers without managing servers or clusters, there's another new service called AWS Fargate, Jassy said.
SEE: Cloud computing policy template (Tech Pro Research)
On the database side of things, Jassy started with AWS Aurora, announcing in preview Aurora Multi-Master. This creates multiple reads, writes, and nodes across multiple availability zones (AZs). Jassy claimed that Aurora Multi-Master will allow for zero app downtime from any node or AZ failure and faster write performance.
A new serverless option for Aurora—simply called Auror Serverless—was also announced at re:Invent. According to Jassy, it's a serverless database that is on-demand and provides automatic scaling as needed.
Two new features came to DynamoDB as well. DynamoDB Global Tables is the first fully-managed, multi-master, multi-region database in the world, Jassy said. DynamoDB Global Tables is generally available today.
DynamoDB Backup and Restore is another new feature that will allow for on-demand, continuous backups along with point-in-time restores for the last 35 days, Jassy said. This could help with application errors or data corruption, and will be available in 2018.
Rounding out the database news was Amazon Neptune, a fully-managed graph database in the cloud, which is now available in preview.
Compute is a huge part of the AWS ecosystem. At a Tuesday night keynote for the conference, the firm unveiled a host of new EC2 compute instances including H1 instances for big data M5 instances for general purpose compute, and EC2 I3 Bare Metal instances.
"EC2 I3 Bare Metal instances give applications direct access to Intel Xeon E5-2686 v4 processors, 512 GiB of memory, 36 hyperthreaded cores, and 15.2 TB of local, SSD-based NVME storage," an AWS press release noted.
AWS also unveiled managed Amazon MQ, a message broker service for Apache ActiveMQ that makes it easier for customers to build and manage messaging services for their applications and services. It integrates with AWS Identity and Access Management (IAM) for authentication and CloudWatch for metrics and analytics.
For AWS EC2 Spot, AWS also brought hibernation for Spot instances and simplified Spot pricing as well.
Another new compute feature is Array Jobs for AWS Batch, announced in a press release. Just as it sounds, this feature allows an array of jobs to be submitted with a single API call, the release noted.
For storage, the new S3 Select feature allow you to pull out only the object data you need using standard SQL expressions, and Glacier select allows customers to run queries on data stored in Glacier.
The 3 big takeaways for TechRepublic readers
- AWS got a big container boost at re:Invent, with EKS for Kubernetes clusters and Fargate for deploying containers without managing servers or clusters.
- Amazon also updated its Aurora and DynamoDB database with new features and announced a new graph database called Neptune.
- A host of new EC2 compute instances and a new messaging system were also announced at re:Invent.
- Special report: The cloud v. data center decision (free PDF) (TechRepublic)
- AWS re:Invent 2017: PostgreSQL on Aurora headlines pre-show announcements (ZDNet)
- Learn Cloud Computing with AWS (TechRepublic Academy)
- AWS announces public preview of EC2 Bare Metal Instances (ZDNet)
- 5 steps for a successful large-scale cloud migration to AWS (TechRepublic)