Share on facebook
Share on twitter
Share on linkedin

5 takeaways from Andy Jassy’s big re:Invent keynote

Forrest Brazeal
Forrest Brazeal

OK, that was … a lot.

Any expectations that AWS would pace themselves throughout this three-week re:Invent marathon went away when Andy Jassy started flinging new service announcements like early December snowballs. Or is that Snowballs?

We counted more than 30 major pause-for-applause announcements throughout Jassy’s three-hour session, with at least six new top-level named service reveals. As usual, while not every new AWS announcement is relevant to everyone, there was certainly something here for each AWS customer.

Here’s a rundown of five key takeaways from the most concentrated cloud announcements firehose of the year:

AWS flipped over the serverless table … again

Just when you thought you more or less knew AWS’s serverless capabilities, they push the limits again – rolling out per-millisecond billing for AWS Lambda, 10 GB of memory and 6 vCPUs for functions, and a whole lot more.

Container image support for Lambda is probably the grabbiest headline here — it’s something customers have been asking for basically since Lambda’s initial release, and while the move away from a code-level abstraction muddles some of the purity of Lambda’s original FaaS concept, it’s clear that containers are the abstraction most developers are comfortable building and tooling on.

It won’t get as much attention, but don’t sleep on the “reinvented” Aurora Serverless SQL either. The first version of Aurora Serverless left a lot to be desired, but the idea of a scale-to-zero, HTTP-based SQL database remains utterly compelling, and if AWS has cracked the cold start issue this feature could end up having as big an impact for builders as anything else announced this year.

To tie all that together, we got AWS Proton – a fully managed deployment service for containers and serverless apps. Notice how those two concepts are starting to blend together? Watch for that trend to continue in the announcements to come.

AWS just keeps getting better and cheaper

I always used to bring up DynamoDB On-Demand Capacity as an example of how the cloud improves under the hood at no cost to customers. Lately, I’ve started to mention AWS Graviton more in that context. AWS’s chip-level innovations let them provide faster, cheaper EC2 instances as first-class citizens in the console, just a click away – no re-architecture required. This is a level of innovation totally beyond what individual customers could build on their own. You can only get this kind of benefit at cloud scale.

Add in per-millisecond billing for Lambda and the new “Trainium” ML chips, and AWS’s economies of scale keep paying dividends for all of us.

AWS continues to innovate on their core services

EC2 and EBS aren’t the new shiny thing anymore, but they still power the vast majority of AWS workloads, and AWS has announced serious improvements to both. (No announcements for fellow workhorse S3, but there’s lots of conference to come yet.)

The separation of IOPs from storage in the new gp3 EBS volumes is a huge win for cost and flexibility. And io2 Block Express just cranks the already ridiculous throughput of the io2 family up to even bigger extremes.

And it was actually announced last night, but don’t sleep on EC2 instances for Mac. The pricing’s a bit wonky, but Apple developers will find ways to work with it to keep their CI/CD in the AWS ecosystem.

Doubling down on open-source and hybrid

Jassy made great points about the evolution of hybrid cloud as a concept, something we’re also hearing from our community here at ACG — no longer a transitional placeholder for customers working their way from on-premises to the cloud, hybrid is now about building a unified set of best practices connecting edge devices to central cloud management.

ECS and EKS Anywhere, as well as open-sourcing EKS itself, give AWS a coherent container management story at all levels of the hybrid relationship. And with new, smaller, AWS Outpost sizes, as well as the promise of more than a dozen new Local Zones, it’s clear AWS wants to be a hybrid partner for everyone … not just giants with bespoke use cases.

Today’s announcements went deep, not wide

Despite several new top-level service announcements – I’m certainly intrigued by Monitron, Lookout, and Panorama, edge-related services that I’ll probably never have a reason to play with — the vast majority of today’s reveals were about upgrading, solidifying, and doubling down on themes and services of re:Invents past. More features for Sagemaker and Connect! Better usability for containers and serverless!

That’s the power of AWS, of course: even if individual announcements underwhelm, you know the service teams will continue to listen and improve over time. It’s the power of customer obsession, and it’s the only thing you can safely assume will remain true about AWS.

Don’t forget to throw in your email address to get an executive summary in your inbox from ACG each day of re:Invent!


Get more insights, news, and assorted awesomeness around all things cloud learning.

Sign In
Welcome Back!

Psst…this one if you’ve been moved to ACG!

Get Started
Who’s going to be learning?