After three weeks packed with announcements, AWS re:Invent wrapped this past Friday.
This year’s event was entirely online, due to the COVID pandemic. And like most online events, it was pretty tiring and boring at times. After all, filling three weeks with content isn’t easy by any means.
At the same time, there were some MAJOR announcements that everyone working with AWS should know.
Below, you’ll find what I think are the most important news that came out of this years’ re:Invent.
Let’s get started.
In one of the biggest announcements this year, EC2 instances are finally available for macOS.
There’s not a lot else to say here.
Of course, not it’s not like everyone’s gonna start using these new instances. But for Apple developers, this is simply a game-changer.
Proton is a new service that’s still in preview. It’s fascinating and confusing at the same time.
First, let’s get something out of the way:
If you’re a small organization that doesn’t have a central cloud/infrastructure team separate from the dev team, Proton is NOT for you.
Also, if you’re just starting out with AWS, leave this one for later. It’s pretty complex and most of what it does is useful only at the enterprise level.
So, what exactly does Proton do? Well, here’s AWS’s definition:
“Today we are introducing AWS Proton: a service designed for the platform engineering teams who want to offer their own self-service interface that provides opinionated methods for running serverless and container-based applications on AWS.”
Clear as day.
In human terms, this is a service for cloud/infrastructure/platform teams that support multiple development teams. With Proton, the central cloud team can package up environments and send them over to the dev teams.
It’s main goal is to solve the problem of app deployment for large organizations. Like AWS puts it:
“With AWS Proton, the service team no longer needs to learn a new Infrastructure as Code (IaC) language, or understand container orchestration, build pipelines, or autoscaling logic. They can focus on writing code.”
There’s much more we haven’t yet learned about this service as it’s still in preview. But one thing is certain – it’s the most fascinating and confusing service AWS has put out this year.
Further reading: AWS Proton is Conway’s Law-as-a-Service
DevOps Guru is a new service (still in preview) that injects Machine Learning into performance and availability monitoring.
ML models can be really powerful for detecting operational issues. The problem is, you have to know how to use them. Obviously, most DevOps engineers aren’t ML experts, which is where DevOps Guru comes in.
Here’s how it works:
You give it an AWS account (or CloudFormation stack) and operational data about it. From here, DevOps Guru works its ML magic to:
- Analyze streams of data
- Monitor relevant metrics
- Establish normal application patterns
The benefit of this approach is that it reduces the time and effort it takes to detect and solve operational issues.
Further reading: Amazon DevOps Guru Service Page
This is another huge announcement that doesn’t need much context.
The new gp3 SSD volumes for Amazon EBS are cheaper and easier to use than the older gp2 volumes.
With gp3, you get a consistent performance of 3,000 IOPS at any volume size. Plus, you can provision more IOPS without having to add more capacity.
Further reading: Amazon EBS General Purpose Volumes
Keeping up with the monitoring and operations announcements, EC2 now has five new metrics to help you dissect network performance.
Before this addition, you could only monitor network bytes and packets utilization. The new metrics will certainly help with troubleshooting and benchmarking performance.
A great addition that lots of people have been requesting for years.
Further reading: Monitoring network performance for Linux instances
About. Damn. Time.
But that’s all in the past now.
Last week, AWS CloudShell was announced. As most people expected, it’s a web-based shell, based on Amazon Linux 2.
All of AWS’s command-line tools come pre-installed on the shell. There are also lots of pre-configured SDKs.
The shell comes with 1GB of persistent storage per region. And there’s no additional cost for using CloudShell – you only pay for the resources you use through the shell.
Further reading: Command-Line Access to AWS Resources
This is the final major service announcement I’m covering here. And it’s another direct hit to existing GCP and Azure services.
Amazon Location Service lets developers add location data to their app, similar to Google Maps Platform. Obviously, this has major implications for lots of companies, as they now have another alternative to choose from.
And since Google increased the price of its Maps APIs as much as 14x two years ago, many companies have been looking for other options.
It’s too early to tell who’ll win this battle as Amazon Location Service is still in preview. Expect fireworks over the next few years as the location data market gets more and more competitive.
Further reading: Add Maps and Location Awareness to Your Applications
Talks, talks and more talks
AWS put an insane amount of content out for this years’ re:Invent. I haven’t watched all of it and I doubt you will too.
However, if you’re in the mood for a few educational/promo videos, here are my three favorite talks from this years’ re:Invent:
Keynote with Andy Jassy: The ideal video if you want a relatively short recap of the most important service announcements.
Developer Keynote with Dr. Werner Vogels: A deeper dive into AWS CloudShell, AWS Fault Injection Simulator and a few other services.
Infrastructure keynote with Peter DeSantis: An in-depth look at how and why AWS thinks differently about performance and reliability.
Like always, re:Invent gave us the biggest service announcements of the year. My guess is that there won’t be any more big news until mid to late January.
If you want to dive even deeper into this years’ biggest cloud computing event, check out this AWS article.
Finally, if you’re planning to start or expand your company’s cloud computing journey next year, don’t hesitate to contact us.