Welcome to issue number 38 of Last Week in AWS.
The year is winding down, and with it the flurry of AWS announcements that have dominated the past few months. Have a happy holiday season, wherever you are.
This week’s issue is sponsored by CloudHealth.
Reserved Instances (RIs) can appear complicated, but this eBook will ease that. Learn all about: 1) how to make effective RI purchases, 2) new instance types and general usage, 3) planning, managing, and optimizing your purchases, and 4) modifying existing reservations.
A dive into how Lambda concurrency works, and how to think about it.
The official AWS Labs GitHub has a writeup on safe deployment models for Lambda.
Ben Kehoe gives some thoughts on Blue/Green Deployment. This has sparked some interesting conversations around whether or not to treat code and infrastructure separately in a serverless world.
A terrific writeup from Intercom’s Director of Engineering on how to get the most from re:Invent next year. Bookmark it– you’ll want it handy in about eleven months.
A curated list of all of the re:Invent announcements that touch upon serverless.
IOpipe’s tool for analyzing whether your EC2 workload is a candidate for Lambda migrations lives at the wonderfully named Servers.lol.
Choice Cuts From the AWS Blog
Amazon Aurora with MySQL Compatibility Natively Supports Synchronous Invocation of AWS Lambda Functions – Using Lambda functions as stored procedures in a database can now be done synchronously. I’ll have more thoughts on this once I’ve finished screaming in terror at the idea of arbitrary Lambda functions serving as blocking database calls.
Amazon CloudWatch now supports two new chart visualization options in metrics and dashboards– They may not be Brendan Gregg’s flame giraffes, but these visualizations definitely bring added clarity to CloudWatch metrics.
Amazon ECS Supports Memory and CPU Limits at the Task Level – A capability enhancement as ECS bids adieu to the “unlimited” adjective, to the chagrin of Marketing.
AWS Direct Connect brings the Amazon network to South Africa in Cape Town & Johannesburg – “Not including Antarctica, we’ve got a presence on all six continents.” “There are seven continents, Timothy.” “I don’t think that’s true.” It turns out that Direct Connect is the fastest AWS service to stand up in a blind panic.
AWS Server Migration Service is now available in US West (Northern California) Region – It’s nice to see that the Server Migration Service has itself migrated west for the winter.
EFS File Sync Now Available in All Amazon EFS Regions – If you have a syncing feeling, rest assured it’s not just you.
Introducing Amazon Linux 2 – It’s the Amazon Linux you’ve come to know and tolerate, now with systemd and a host of other changes that definitely won’t break your application stack at all. Maybe let other folks find the sharp edges first on this one?
Introducing Field-Level Encryption on Amazon CloudFront – CloudFront continues to be outstanding in its field… level encryption.
How to Manage Amazon GuardDuty Security Findings Across Multiple Accounts | AWS Security Blog – If the S3 Bucket Negligence awards are anything to go by, everybody stopped reading this one at the word “security.”
Introducing the New GDPR Center and “Navigating GDPR Compliance on AWS” Whitepaper | AWS Security Blog – If you do business in Europe and haven’t heard of GDPR yet, stop reading immediately, pour yourself a strong drink, and read this. CONSULTANTS ARE STANDING BY!
How Uber survives its busiest nights of the year | AWS Startups Blog – Oof. Fascinating technology story here, but they could really have picked a better time– or company to highlight.
S3Guard brings consistency to Hadoop when backed with S3.
Have a lambda function that iterates through all of your regions and terminates any resource that lacks a user provided tag.
If you need a spot instance quickly and don’t want to spend a lot of time getting one up and running, take a look at py-spotme.
…and that’s what happened Last Week in AWS.