Energetiq @ re:Invent 2018
At Energetiq we pride ourselves on staying at the forefront of the services that AWS provides. So naturally when Amazon hosts a massive conference called re:Invent in Las Vegas with 52,000 attendees, announcing all sorts of new AWS services and updates we were always going to be right in the middle of it.
The first thing that most attendees find confronting is the scale of the event. Spread across 7 hotels and with over 3500 sessions to participate in, even moving around the conference requires logistics few organisations could muster.
In amongst the sessions were several keynotes by some very influential people within AWS. People such as Andy Jassy (CEO of AWS), Peter DeSantis (VP, AWS Global Infrastructure) and finally Dr Werner Vogels (CTO and head of Innovation). When you consider that AWS is now the dominant provider in the IaaS market, being able to get insights into the road ahead for AWS from some of their top people was an opportunity not to be missed.
But the one thing everyone wants to take away from a conference such as re:Invent is what exciting new technology and services are just around the corner. So let’s get into it.
More speed, more power
There were several announcements on the ways AWS are enabling faster capabilities and the first of these is their new Global Accelerator service. This service leverages the AWS network to provide a highly available, low congestion pipeline for customers running their AWS infrastructure in multiple regions.
But an even bigger announcement is the news that AWS are making available a new instance type for compute heavy applications and utilising up to 100 Gbps of networking throughput. We’re pretty excited about this as many of our applications are already processing large amounts of batch type data and indications are we will be able to deliver more speed just with a few clicks in the AWS console.
More Lambda, less pets
So whilst we all love hearing about speed and power increases, those of us who have been responsible for maintaining EC2’s or other server types are likely already on the road to Serverless. AWS enables this mindset in many ways but the key contributor to this approach is AWS Lambda providing the ability to run your application code ‘somewhere’ on the cloud and letting AWS provide the maintenance for the server running your code. AWS have continued to build on the Lambda service in the following ways:
Lambda layers — Use code and other objects from external libraries without having to bundle into your deployment package.
Support for Ruby and Customs runtimes — Run Ruby or custom runtimes as a Lambda function
IDE Support — There is now a greater range of IDE support for Lambda
ALB Support — The Application Load Balancer (ALB) can now invoke Lambda functions.
These services enhance an already powerful tool but we’re excited by these because it effectively represents a maturing of the service to align with some of the best development practices and to integrate into your organisations development framework.
Data, Data, everywhere
The concept of data lakes are not exactly new and AWS have been consistently making these services easier to assemble with services such as S3, Redshift, Glue and Elastic Map Reduce (EMR). But you are still effectively building your data lake from these pieces and depending on the variety of data sources, this can be a time consuming exercise. At re:Invent, AWS have announced an exciting new service called Data Lake formation. The service brings together all the building blocks already provided into one centralised place to streamline the assembly of your Data Lake. Data lakes are a growing trend across the industry and this new feature is one we feel is worth paying close attention to.
Machine learning for all
Machine learning has been a growing sector over the last few years and re:Invent showed us that continues to be the case. Services such as Sagemaker are growing in popularity, but there is still is a steep learning curve to be able to use machine language effectively. So the big news from AWS is now that they are now integrating machine learning directly into some of their services to enhance those services. And best of all you don’t need to be an expert in machine learning to take advantage. An example of this is the addition of machine learning to the QuickSight product to drive anomaly detection and forecasting with just a few clicks. This enables non-technical users to start seeing the benefits machine learning can provide without the need to understand what can be complex machine learning models.
Get building
Continuous innovation is a key philosophy for AWS and re:Invent has once again shown there are many new services and capabilities to support construction of new products and services. Time to get building!
Resources
https://aws.amazon.com/global-accelerator/
https://aws.amazon.com/lake-formation/
https://aws.amazon.com/blogs/big-data/amazon-quicksight-announces-ml-insights-in-preview/
