2020 has definitely been a rollercoaster of a year and AWS had to adjust this year's re:Invent, taking it completely virtual. And the AWS team behind re:Invent has not disappointed. After a less than stellar experience with the summits earlier in the year, I was interested to see how re:Invent would go. With the exception of the poor scheduling tool, I managed to book Andy's keynote into my calendar and had no one book a meeting over it. So I kicked back with a coffee and enjoyed this year's re:Invent keynote from the comfort of my home office.

The theme (as always) was about reinvention. What does it take to reinvent yourself as a company? Andy started off by stating that in time to come we'll look back at this year and see how the pandemic accelerated cloud adoption by a number of years. He has seen that the company's who were already in the cloud were able to adapt quickly to cope with varying workloads (in some instances that was up, in others, it was down). The key message is "don't wait too long to reinvent".

Andy then asked the question, "what does it take to reinvent?", and listed 8 key areas that companies should adopt.

  1. Leadership will to invent and reinvent.
  2. Acknowledge that you can't fight gravity.
  3. Talent that's hungry to invent.
  4. Solve real customer problems.
  5. Speed.
  6. Don't Complexify.
  7. Use the platform with the most capabilities & broadest set of tools.
  8. Pull it all together with aggressive top-down goals.

I think Andy is on to something here and in my experience I've also seen many of these traits in high-performing organisations. To sum up Andy's words, continuous reinvention is the key to building a sustainable business. Organisations should always be reinventing and developing products and services that solve customer problems to thrive in this competitive world.

Some of the announcement highlights from Andy that I think were interesting are:

Compute

  • c6gn instances - A lot of people have been experiencing some excellent price/performance improvements with the Graviton processes.
  • New Intel Habana Gaudi-based Amazon EC2 instances - 40% better price/performance over current GP based EC2 ML instances.
  • New AWS Trainium - More new silicon from Amazon/AWS, this time specialising in Machine Learning.

Serverless

  • Amazon ECS Anywhere - Allows you to run ECS on-premises
  • Amazon EKS Anywhere - Allows you to run EKS on-premises
  • Lambda Container Support - Using containers as your deployable artefact for Lambda instead of a zip file.
  • AWS Proton - A fully managed deployment pipeline for containers and serverless

Other

  • GP3 storage - this lets you tune both size and performance for the right size/performance that you need
  • io2 block express volumes - this storage type is now really similar to traditional SAN, and will have new features added during 2021
  • Amazon Aurora Serverless v2 - Brings all the learnings from the first generation of Aurora Serverless into this 2nd generation for an improved experience
  • Babelfish - TSQL to PostgreSQL on the fly translator
  • AWS Glue Elastic Views - Cross datastores materialised views
  • Amazon DevOps Guru - An AI-powered service that helps to identify potential operational issues and suggest improvements.

And many, many more.

Andy left us with one of his favourite quotes:

"We want missionaries, not mercenaries — people focused on building businesses that last beyond their tenure at the company".