First timer at AWS re:Invent

AWS re:Invent is a yearly learning conference by Amazon Web Services for global cloud community. This year it was held in December 2nd – December 6th. 65 000 people around the world gathered in Las Vegas for this week long event for networking, sessions, workshops, labs, beers and much more.

 

Six casinos in the middle of the Strip were reserved for re:Invent sessions and another six were sleeping venues. Session venues hosted about 3000 sessions covering every imaginable AWS offering and lots of case studies.

Getting ready

Seats for sessions were opened for reservation a while before the event. I had a long list of interests and I was able to reserve a seat for almost all of them. Only one or two were full when I tried. Fortunately, they were saving some seats for walk ups too. So everything was set, I was ready to go.

First Impressions

IMG_20191201_173543

Las Vegas is a familiar place for me. This was my 4th time in the city so I knew pretty much what to expect. I knew what my hotel/casino was like, where it was located and how to travel between sites.

The conference, on the other hand, was a different story. The settings were huge. I was really impressed. The whole Strip and all casinos were full of people in yellow AWS shirts ready to help you with anything. Chance of getting lost was below zero. Got my badge in no time and was ready for the overload of knowledge!

The conference

First thing Monday morning was a cloud migration workshop. Me, one British and a few Americans did a roadmap for AWS cloud migration. Straight into deep end with my jet lagged head. Good fun!

Most of my interests were chosen from leadership perspective and not so much from technical. And lot of Machine Learning. ML was visible everywhere. Algorithms, scaling, DeepRacer and the big ML announcement: SageMaker Studio with all the bells and whistles. It’s a ML Integrated Development Environment (IDE) including for example Debugger, Autopilot, Experiments and Model Monitor that lets you easily build, train, debug, deploy and monitor your machine learning models. Read more about SageMaker Studio here: https://aws.amazon.com/blogs/aws/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

DeepRacer was also everywhere. It is a 1/18th scale self driving miniature car. It learns how to drive using reinforcement learning that is easy and fun to train. Then you compete against each other about who has the best model. There is no need for deep technical knowledge, one of the best “drivers” has been a designer from Japan with no technical background. The conference had an arena with 6 tracks for DeepRacer League finals.

IMG_20191203_135507

Nothing about Serverless? YES THERE WAS! And a lot! Best practices, architectures, monitoring and observability, security, talks and workshops. Serverless really is the next (and current) big thing in computing. Many new announcements like Provisioned Concurrency (https://aws.amazon.com/blogs/aws/new-provisioned-concurrency-for-lambda-functions/) to provide greater control over the performance of their serverless applications. This will, for example, help with cold start issues.

Another subject that caught my attention was serverless data streams and analytics. It’s fun to realize how straight forward tools and workflows they have in place for such use cases. The Kinesis product family with S3 will take you far, and getting started with data lakes couldn’t be simpler.

Some other interesting announcement were Outposts for hybrid cloud, Builders’ Library that takes you under the hood how Amazon works with AWS, Fargate for EKS and CodeGuru for automated code reviews.

Every night was loaded with side events from sponsors as well as from AWS. There was no shortage of things to do, places to go, and no need be alone. Or thirsty. Even sumo wrestling was available.

Lessons Learned

I had my calendar full of program. But what I did not think of enough was the location of the sessions. I knew the distances in Vegas are big but I did not take into account that I might be a bit exhausted after couple of days. So I wanted to make changes into my schedule so that I wouldn’t have to travel between casinos too much. Fortunately that worked out well.

Save some time for networking, exhibitions and other events. Re:Invent is not just about sitting in sessions, it’s as much about networking, experiencing, experimenting and fun. I would leave empty slots in my calendar for freestyling. Many of the new announcements during the conference will have their own sessions and workshops, which also are scheduled after the announcements. So be prepared to fit them in your schedule!

The event was huge. And I mean enormous. But arrangements worked like a dream. For example I never had to wait to get lunch or transportation. Just walk right in, get your food and eat. That is really remarkable for an event for 65 000 people. Would I go again? Definitely but I would like to take somebody with me. Discussing the things seen and learned daily with somebody would add lot of value.

And shoes. Take the comfortable pair.

 

Written by Jari Rantala

What's next

See our latest projects

Microsoft – Innovative Analysis of Market Research Data

Microsoft – Innovative Analysis of Market Research Data

Even though market research is not a new thing, an innovative and diverse analysis of market research data in the way Microsoft’s Data & Analytics Platform team has done it, is unique both in Finland and globally.

Upcoming events/courses

Scrum Master | Helsinki 20.-21.8.2020

Scrum Master

Helsinki 20.-21.8.2020

The Scrum Master course is a two-day course which introduces the concepts of agility using the Scrum framework. Course can serve as a preparation for certification assessments, e.g. the Professional Scrum Master™.
Sign up

Subscribe to Stories