I’ll be giving a new talk tomorrow evening at a community meetup for LOPSA ETENN. It’s online so you should attend; I’m calling it You Might Experience Some Discomfort: Your New AWS Account & You. I promise nothing other than that it’s likely to be amusing.
Cloudflare set the internet abuzz with its new R2 object store; this has the potential to effectively wipe out egress charges for S3 data. I’ll have more on that this Wednesday, but to summarize: if it functions as described, using Cloudflare’s D2 for frequently read S3 objects won’t just be a best practice, but rather “you’ll be negligent if you don’t do it.”
From the Community
Think the data lake is dead? Well think again – because with the ChaosSearch Data Lake Platform – the data lake is back! And now, courtesy of ChaosSearch, get complimentary access to the new 2021 Gartner Hype Cycle™ for Data Management. This new Gartner report assesses more than 30 different categories of data management technologies — including data lakes, multi-model DBMS, logical data warehouses, and more! Get your copy of this new Gartner report to learn: The top data management technologies in use today; which vendors offer solutions for each category of data management tech, and more! Take it from me, Corey Quinn, or take it from the growing list of ChaosSearch customers like Klarna, Blackboard, Equifax, Armor and more… the data lake is back.
I’ve been a fan of the Oracle Cloud free tier for a while (which is a very different thing than being a fan of Oracle itself, I assure you). Here’s an analysis about some of its quirks. I want to be very clear here: those quirks are around its behaviors, and some of the decisions made in how it’s constructed. There is no “quirk” that pulls an AWS Free Tier and charges you a few tens of thousands because you forgot to check a box somewhere. You must affirmatively upgrade your account before they will charge you a single cent. I really can’t overstress how important that is.
Amazon The Open Source Company Who Truly Gets Community demonstrates it by completely ignoring a pull request to one of their open source projects for five months, then coming back with a “nah, we’d have to maintain this and that takes resources we don’t have.” Do we need to throw the $1.7 trillion company a bake sale or something?
RedMonk’s Rachel Stephens has a great review of Cloudflare’s new R2 service, and A New Strategy, R2 is just a perfect title for it.
If you had told me a year ago that I’d take what an AWS employee had to say about the future of DevOps and software development seriously, I’d have scoffed and asked whether they were planning on hiring Emily Freeman. Well, they did and now I do. It’s worth a read and watch.
A breaking and unpublicized change in how CloudTrail will present Lambda executions means that some of your detection rules may stop working. You probably want to check those.
AWS acquired Stackery. Okay, that’s great. Companies acquire other companies all the time; it’s common. But the only notice of this as of this writing is in an email sent to existing Stackery users and customers telling us that the service was being turned off. There has been no formal notice from AWS, and let’s face it: I’d reach out to AWS for an explanation but I already know that an AWS Spokesperson Will Decline to Comment®. Best of luck to my (now formerly) Stackery friends.
Azure has had some serious challenges lately, across a variety of fronts. Alex Hudson sums them up neatly in his post it’s tough being an Azure fan.
My post about where I see The Actual Next Million Cloud Customers coming from seems to have struck a broad unspoken consensus. There’s something here…
Protocol has a piece out about how IBM Cloud lost the cloud war. True, the corpse hasn’t stopped moving around and squawking about hybrid cloud – but it’s dead. The tragedy here is that some of the smartest people in the world work at IBM, and they’ve been restrained from fixing IBM Cloud’s issues. The capacity and talent are definitely there, but the leadership and executive will are absent.
If you’ve got an interesting job for this newsletter’s eminently employable subscribers, get in touch!
The Duckbill Group (that’s me!) is hiring a Head of Consulting Services to join the team. We’re looking for someone skilled in managing and leading people, as well as in building and optimizing delivery processes. As a member of the leadership team in a nine-person company, you contributions will be instrumental to our continued growth and success. AWS expertise isn’t required, but it’s certainly a bonus. If you’re interested in a role that’s fully-remote, has big impact, and you want off the VC rollercoaster, come check us out.
Trajectory Conference is One Giant Leap for DevOps
Software powers the world. LaunchDarkly empowers all teams to deliver and control their software. DevOps and feature management are reimagining how businesses build and release new products. On November 9th and 10th, LaunchDarkly is hosting Trajectory Conference 2021 — a two-day event for software innovators who want to break orbit, not systems. Trajectory is a fully-virtual conference that focuses on the technology, people, and processes that continuously deliver better user experiences and more powerful software. Register today to join fellow thought leaders shaping the future of software development.
Amazon ECR Public adds the ability to launch containers directly to AWS App Runner – Excellent, another way to run containers on AWS. They need a few more of those.
Amazon Monitron launches iOS app – I don’t have Monitron devices but I installed the app anyway. It doesn’t let me do anything without a Monitron device, so I’m leaving it a 1 star review on the app store and a review saying “this game sucks.”
Now auto-terminate idle EMR clusters to lower cost – I don’t know as I’d classify “turn the cluster off when it’s done with its job” as a particularly innovative feature so much as I would one that’s very long overdue.
AWS announces AWS Snowcone SSD – It’s a bit more expensive, it holds a lot more data, it’s worlds faster to access its disks, and it still has battery life of roughly half a second, so I’m not at all impressed with this new Amazon Kindle.
AWS announces the general availability of AWS Cloud Control API – Once this gets more global coverage I’ll look into it. Right now it feels like it’s more for tool builders than tools themselves. I await emails correcting me.
AWS Backup simplifies recovery point deletions – It’s now way easier to accidentally delete that backup upon which your company depends.
Monitoring clock accuracy on AWS Fargate with Amazon ECS – I have a simple rule: if AWS does an announcement that touches Amazon Time Sync Service, I write about it here. I love that thing; ntp is an old hobby of mine.
Improving Performance and Reducing Cost Using Availability Zone Affinity – …because AWS of course charges you 2¢ per GB that moves between AZs, forcing you to decide just what a reliable, durable architecture is really worth to you.
Amazon Managed Service for Prometheus Is Now Generally Available with Alert Manager and Ruler – “Manager and Ruler” is probably the most egregious case of title inflation I’ve seen so far; AWS’s attrition issues might be worse than I’d thought.
AWS Lambda Functions Powered by AWS Graviton2 Processor – Run Your Functions on Arm and Get Up to 34% Better Price Performance – This is handy both for ensuring that my Arm development environment is now on the same architecture as the Lambda environment is, as well as for the small handful of customers with Lambda bills big enough to care about.
New – Amazon Genomics CLI Is Now Open Source and Generally Available – …and another way to run containers on AWS, no doubt.
Now — AWS Step Functions Supports 200 AWS Services To Enable Easier Workflow Automation – There used to be 17 ways to run containers in AWS, and 17 AWS services supported by AWS Step Functions. Now there are 200 of at least one of those.
Coming January 2022: An updated Amazon QuickSight sign-in experience – I’m struggling to reconcile “will undergo minor updates” with “three months’ notice” here…
Defining an AWS Multi-Account Strategy for a Digital Bank – Wells Fargo probably wouldn’t have picked both Azure and Google Cloud as their primary cloud providers if AWS had included the term “multi-account strategy” in their proposal.
Complying with city-level embargos using Amazon CloudFront – I confess I didn’t realize that there were embargoes so narrowly scoped that you could evade them simply by going for a walk.
AWS in Virginia: The economic impact of solar – The linked PDF states that from 2011-2020, AWS spent $35 billion on data centers in Virginia. That’s a huge pile of money, and really speaks not only to the sheer scale of what AWS does, but also to just how much money they’re making in Managed NAT Gateway data transfer.
Wickr, an AWS company, offers a secure and compliant solution to protect organizational communications – Yes, why fix Chime when you can simply acquire a similar service and further confuse the messaging. This is very Amazonian.
Observability is critical for managing and improving complex business-critical systems. With observability, any software engineering team can gain a deeper understanding of system performance, so you can perform ongoing maintenance and ship the features your customers need. Preview Honeycomb’s upcoming O’Reilly book to understand the value of observable systems and how to build an observability-driven development practice.
aws-auth is a neat take on login flow. It’s similar to AWS SSO or aws-vault, but a lot more modern-feeling, and becomes significantly more granular.
Somehow I never encountered placebo before; it lets you record the results of API calls to AWS so you can replay them for things like unit testing. It’s both ridiculous and true that “taking a placebo makes your AWS bill feel better.”
I’d forgotten about rclone! It’s effectively “rsync that speaks to cloud storage natively.” Super handy if you need it.
… and that’s what happened Last Week in AWS.