"What happens in Vegas, stays in Vegas". As we found out, this is not always true. There was much for us to take back with us from the AWS re:Invent conference in Las Vegas.
Take 65,000 participants, over 3000 sessions across different venues in only a couple of days and several various events, without anybody going hungry – these are just some of the things that portray the sheer scale of this conference. The flawless logistics justifies the claim of this being one of the biggest conferences in the world. Also, it is Las Vegas which make this an experience to remember for life.
A CONFERENCE FULL OF INNOVATION
Hopping from one session to another made us realize the amount of innovation going on at AWS. Despite the fact that AWS is the leader when it comes to cloud vendors, bigger than the next four largest vendors combined by market share, the drive for innovation going on at AWS is impressive. Despite being number one, there is no notion of a "big company syndrome" - the innovation malaise often prevalent in large corporations. AWS has none of that syndrome and is constantly improving already successful services and releasing new ones. This is generally speaking a trait of successful companies and Andy Jassy, current CEO, "summarized" only some of these innovations in his three hour long Keynote speech.
Without doing a full taxonomy of what is new in AWS, here are a few that we found intriguing with respect to our current projects:
AWS CodeGuru - Code analysis tool that makes use of machine learning to find the most expensive lines of code that hurt application performance. Also, it proposes recommendations to fix or improve the code. Currently it supports Java only, but more languages are to come and we can't wait to start experimenting with this tool. This being a new tool, we have not yet had the chance to evaluate this service, but we do have high hopes.
AWS EKS on AWS Fargate - Why is this important? Well, this reduces the maintenance of infrastructure when using EKS based on serverless technology of Fargate.
AQUA (Advanced Query Accelerator) for AWS Redshift - Despite the huge success of Redshift as the data warehouse storage and computational engine, AQUA improves query performance ten-fold.
AWS SageMaker - There are many new additions to the SageMaker ecosystem, namely Studio, Notebooks, Experiments, Debugger, Model Monitor and Autopilot. All of these simplify machine learning on AWS. Having used machine learning and understanding the difficulty of finding the best hyper-parameters during training, the Autopilot is great addition as it mitigates much of the burden in making these choices. It trains and tunes the best machine learning models based on the data by automatically exploring different solutions to find the best model.
To summarize the above into three words: learn, improve and innovate.