Good Morning!
My re:invent talk has been posted; go check it out and marvel at my ability to make Cloud Financial Management (motto: "Because watching paint dry is too exciting!") come to life, and become something people plausibly care about.
From the Community
Antirez has returned to work at Redis Labs. Redis, of course, is a proprietary fork of the Valkey key-value store.
RedMonk did a writeup of one of the more interesting cloud companies: Cribl. If that name sounds familiar, it may be because Splunk tried to sue them under premises so ridiculous that the entire case got tossed on summary judgement. I’m sure this has NOTHING WHATSOEVER to do with the fact that using Cribl dramatically cuts the amount of money a company pays Splunk every month…
This came out last year, but it holds up really well: Should you use a Lambda Monolith, aka Lambdalith, for your API?
Wow, it’s rare to see a cybercrime group themselves score an S3 Bucket Negligence Award.
Podcasts
Last Week In AWS: A Return to Greatness, or Degenerate Day 3?
Screaming in the Cloud: Looking at the Current State of Resilience with Spencer Kimball
Screaming in the Cloud: Replay – Multi-Cloud is the Future with Tobi Knaup
Choice Cuts
Amazon Bedrock Guardrails reduces pricing by up to 85% – Hard not to read a price cut this dramatic in under a year since the service launched as either "we massively screwed up the pricing" and/or "we’ve been massively undercut by a competitor."
Amazon CloudWatch now provides centralized visibility into telemetry configurations – Honestly, for most folks centralized visibility comes from using a higher level system above CloudWatch. Think Datadog or similar…
Amazon EC2 F2 instances, featuring up to 8 FPGAs, are generally available – I love these weird instances. They’re a niche that most people don’t realize exists, but the 2nd generation makes it clear that AWS hasn’t forgotten about you.
Amazon SES now offers Global Endpoints for multi-region sending resilience – 5 9’s of uptime for… an email sending service? Seems odd to me; SMTP is a best effort protocol. You can do all the optimization you’d like around resilience, but ultimately your queue builds and mail is delayed whenever the recipient’s mailserver takes a nap.
AWS Toolkit for Visual Studio Code now includes Amazon CloudWatch Logs Live Tail – Thanks, I hate it. There are a bunch of "tail CloudWatch log groups, individually or in aggregate" offerings out there, some from AWS, just ready to go on GitHub. The official Live Tail offering instead, of course, costs customers money.
Accelerate your AWS Graviton adoption with the AWS Graviton Savings Dashboard – I like this. I spoke about it in my re:Invent talk, and it’s real. Switch to Graviton if you can, your budget will thank you for it.
How AWS sales uses Amazon Q Business for customer engagement – I haven’t heard a peep from my AWS account team (which reads as: sales), so that tracks perhaps better than they intended it to.
AWS Network Firewall Geographic IP Filtering launch – What’s the point of blocking some locations from accessing your resources, when VPNs are plentiful and cheap? Compliance, dear reader. Compliance.
Issue with DynamoDB local – CVE-2022-1471 – Something of a nothingburger here. Fix it if you’d like, but this is a dev tool that you can’t run in production without violating terms of service, so it’s not exactly a burning fire.
Tools
Sometimes a tool’s name is Just too good.
… and that’s what happened Last Week in AWS.
