Making sense of Re-Invent Announcements

Making sense of Re-Invent Announcements

AWS re-Invent 2017—from November 27 to December 1—lived up to tradition with a spate of announcements, generating tremendous excitement. Clearly AWS has established itself as the leader in cloud-infrastructure services and these announcements— based on cutting-edge technologies—are largely aimed at wooing developer teams to facilitate creation of relevant business applications, easily, quickly. At Umbrella we debated what the most impactful services were and here is our list of choicest services.

AWS Fargate

This is by far the most potent service that will likely boost container development, enabling developers to simply develop, package and deploy applications without having to manage the availability, capacity and maintenance of underlying infrastructure. As a platform for developing container services, Fargate does not allow customization but it takes flexibility and agility to new heights since it is natively integrated with AWS VPC, CloudWatch, IAM and load balancers, allowing developers to easily get started.


AWS EKS is managed Kubernetes container service facilitating developers to build and orchestrate micro-services without having to manage Kubernetes clusters. Although 63% of Kubernetes workloads run on AWS—according to the Cloud Native Computing Foundation— customers still need to do a lot of manual configuration including installing and operating Kubernetes master and configuring cluster of Kubernetes workers. EKS takes away that challenge, offering built-in redundancy with its multi-AZ architecture. The upside for Amazon EKS is that it runs upstream version of the open-source Kubernetes software and applications running on EKS are fully compatible with those on standard Kubernetes environment—on-premises datacenters or public clouds. This means that you can easily migrate Kubernetes application to Amazon EKS with zero code changes.

Amazon Cloud9

AWS launched Cloud9 an Integrated Development Environment to write, run and debug code, allowing access to rich code editor, integrated debugger and built-in terminal with pre-configured CLI. Cloud9 is pre-packaged with essential tools for popular programming languages including Javascript, Python, PHP, eliminating the need to install various compilers and toolchains. The service provides seamless experience for working with serverless applications allowing to quickly switch between local and remote testing and debugging. AWS Cloud9 provides SSH supports and has powerful pair programming features, designed to make collaborative cloud development easy.

Inter-region VPC Peering

 AWS announced inter-region VPC peering allowing deployments across regions to be connected securely. This means a lot for partners who can simply connect with customer deployments in multiple regions through a secure connection as opposed to building, configuring and managing a VPC connections before. Enterprises can also host applications on AWS Cloud and make it globally available on a secure network. AWS Inter-region VPC peering encrypts inter-region traffic and ensures availability and network performance.

AWS’s focus on Artificial Intelligence and Machine Learning grew sharper with a slew of interesting announcements at re-Invent. Here’s our pick:

Alexa for Business

Alexa for Business is a voice-controlled virtual assistant designed to enable organizations to boost productivity, and build a smart organization by connecting Alexa devices at scale. It can be used on personal and shared devices to increase efficiency and convenience. For example, Alexa can help you join meetings in conference halls by turning on video equipment, dialling into the call and starting the meeting. It can help find useful business information in Salesforce, Splunk or Concur. On a personal level, Alex can help to manage your to-do list, check calendar to schedule appointments and set alarms and notifications.

Amazon SageMaker

Amazon SageMaker is a fully managed end-to-end machine learning service that enables data scientists, developers and ML experts to quickly build, train and host machine learning models at scale. It consists of three main components, each of which can be used in isolation. Build, wherein managed instances for training data exploration are automatically configured and optimized for Apache MXNet and TensorFlow. You can specify the location of your data, type and quantity of EC2 instances based on which SageMaker will provision the cluster and perform training with a single click. Finally the one-click deployment models will auto-scale EC2 instances with HTTPS endpoints for low latency and high throughput inference interfaces.

AWS Rekognition Video

Amazon Rekognition Video is a service that enables analysis of images based on deep learning. It tracks people, detects activities, objects and inappropriate content, and recognizes celebrities, even in live streams. Rekognition Video can analyze video stored in Amazon S3 and return specific labels of activities, people and faces, and objects with time stamps so you can easily locate the scene. Rekognition Video application will have huge ramifications in facilitating public security, generating search index for media and entertainment customers, and monitoring for smart homes, particularly when used alongside Amazon Kinesis Video Streams—a service which streams video from millions of devices.

Amazon Deep Lens

 Deep Lens is a new video camera that is capable of applying deep learning techniques to image and video processing at the local level. This device created a lot of excitement during launch and is bound to garner momentum with the developer community. Deep Lens will likely play a key role in pushing the envelope in applying IoT, edge computing and machine learning techniques in homes and among consumers. A strong scenario of Deep Lens application is strengthening physical security.