Serverless @ re:Invent 2017

At re:Invent 2014, we announced AWS Lambda, what is now the center of the serverless platform at AWS, and helped ignite the trend of companies building serverless applications.

This year, at re:Invent 2017, the topic of serverless was everywhere. We were incredibly excited to see the energy from everyone attending 7 workshops, 15 chalk talks, 20 skills sessions and 27 breakout sessions. Many of these sessions were repeated due to high demand, so we are happy to summarize and provide links to the recordings and slides of these sessions.

Over the course of the week leading up to and then the week of re:Invent, we also had over 15 new features and capabilities across a number of serverless services, including AWS Lambda, Amazon API Gateway, AWS Lambda@Edge, AWS SAM, and the newly announced AWS Serverless Application Repository!

AWS Lambda

AWS Lambda Doubles Maximum Memory Capacity for Lambda Functions – We’ve doubled the maximum memory that you can configure a function to have available, to 3 GB. With this, comes proportional increases to CPU and networking so that your function gets access to two CPU cores!
AWS Lambda Supports Traffic Shifting and Phased Deployments with AWS CodeDeploy – Traffic shifting allows you to deploy your Lambda functions using standard industry best practices such as canaries and blue/green deployments. With CodeDeploy, you get the ability to automate rollbacks and have events fired off by triggers set around the lifecycle of an individual deployment.
- Blog Post: Implementing Canary Deployments of AWS Lambda Functions with Alias Traffic Shifting
AWS Lambda Introduces Enhanced Console Experience – We’ve reworked so many parts of the console that it’s hard to cover it in a single post. Go and check it out for yourself!
Set Concurrency Limits on Individual AWS Lambda Functions – You can now set a concurrency reservation for a function in your account allowing you to limit functions from using too many backend resources or taking up too much concurrency, or to control costs.
- Blog Post: Managing AWS Lambda Function Concurrency
AWS CloudTrail Adds Logging of Execution Activity for AWS Lambda Functions – Gain a better understanding of who or what is invoking your Lambda functions via CloudTrail Data Events for Lambda.
Pre-announcement: Golang and .NET Core 2.0 support coming soon!

Amazon API Gateway

Amazon API Gateway Supports Endpoint Integrations with Private VPCs – You can now provide access to HTTP(S) resources within your VPC without exposing them directly to the public internet. This includes resources available over a VPN or Direct Connect connection!
Amazon API Gateway Supports Canary Release Deployments – You can now use canary release deployments to gradually roll out new APIs. This helps you more safely roll out API changes and limit the blast radius of new deployments.
Amazon API Gateway Supports Access Logging – The access logging feature lets you generate access logs in different formats such as CLF (Common Log Format), JSON, XML, and CSV. The access logs can be fed into your existing analytics or log processing tools so you can perform more in-depth analysis or take action in response to the log data.
Amazon API Gateway Customize Integration Timeouts – You can now set a custom timeout for your API calls as low as 50ms and as high as 29 seconds (the default is 30 seconds).
Amazon API Gateway Supports Generating SDK in Ruby – This is in addition to support for SDKs in Java, JavaScript, Android and iOS (Swift and Objective-C). The SDKs that Amazon API Gateway generates save you development time and come with a number of prebuilt capabilities, such as working with API keys, exponential back, and exception handling.

AWS Serverless Application Repository

Serverless Application Repository is a new service (currently in preview) that aids in the publication, discovery, and deployment of serverless applications. With it you’ll be able to find shared serverless applications that you can launch in your account, while also sharing ones that you’ve created for others to do the same.

AWS Lambda@Edge

Lambda@Edge now supports content-based dynamic origin selection, network calls from viewer events, and advanced response generation. This combination of capabilities greatly increases the use cases for Lambda@Edge, such as allowing you to send requests to different origins based on request information, showing selective content based on authentication, and dynamically watermarking images for each viewer.

AWS SAM

Globals, Safe Lambda Deployments, Lambda Versions & Aliases, Local Lambda execution (GitHub release notes)
Make your apps more secure – SAM Policy Templates (GitHub release notes)

Twitch Launchpad live announcements

Other service announcements

Here are some of the other highlights that you might have missed. We think these could help you make great applications:

AWS re:Invent 2017 sessions

Coming up with the right mix of talks for an event like this can be quite a challenge. The Product, Marketing, and Developer Advocacy teams for Serverless at AWS spent weeks reading through dozens of talk ideas to boil it down to the final list.

From feedback at other AWS events and webinars, we knew that customers were looking for talks that focused on concrete examples of solving problems with serverless, how to perform common tasks such as deployment, CI/CD, monitoring, and troubleshooting, and to see customer and partner examples solving real world problems. To that extent we tried to settle on a good mix based on attendee experience and provide a track full of rich content.

Below are the recordings and slides of breakout sessions from re:Invent 2017. We’ve organized them for those getting started, those who are already beginning to build serverless applications, and the experts out there already running them at scale. Some of the videos and slides haven’t been posted yet, and so we will update this list as they become available.

Find the entire Serverless Track playlist on YouTube.

Talks for specific use cases

Talks from AWS customers & partners

Wipro – Building Smart Applications Leveraging AWS to Drive Customer and Worker Experience (SRV211) (slides)
Dynatrace – Keys to Successfully Monitoring and Optimizing Innovative and Sophisticated Cloud Applications (SRV309)
Homeaway – How We Built a Mission-Critical, Serverless File Processing Pipelines (SRV315)
Agero – How Agero is Preventing and Detecting Vehicle Accidents in Real-time with Serverless Computing (SRV316)
PNNL – Research at PNNL: Powered by AWS (SRV318) (slides)
Nextdoor – How Nextdoor Built a Scalable, Serverless Data Pipeline for Billions of Events per Day (SRV319)
MongoDB – Build a Serverless, Face-Recognizing IoT Security System with Amazon Rekognition and MongoDB Stitch (SRV336)

Looking to get hands-on with Serverless?

At re:Invent, we delivered instructor-led skills sessions to help attendees new to serverless applications get started quickly. The content from these sessions is already online and you can do the hands-on labs yourself!
Build a Serverless web application

Still looking for more?

We also recently completely overhauled the main Serverless landing page for AWS. This includes a new Resources page containing case studies, webinars, whitepapers, customer stories, reference architectures, and even more Getting Started tutorials. Check it out!

AWS Compute Blog