This is a guest post from the (in)famous Corey Quinn, who you may know from such ridiculous things as wearing three-piece suits to tech conferences, having a love affair with dogs that look suspiciously like weasels, and giving some of the tech industry’s most hilarious conference talks.
The conference talk recording of this is available at A Cloud Guru.
There’s been a revolution in the cloud computing space.
In case you’ve been sleeping under a rock for four years, it’s the concept of Functions as a Service. Often referred to as “serverless,” it’s equally often interrupted by some sad sack insisting that “Serverless runs on servers!”
Concurrently, there’s been a holy war between proponents of “monitoring” and “observability.” While a highly nuanced argument, that’s way too much work for my attention span and I doubt I’m alone.
To that end, I suggest a melding of these two pointless arguments into something that everyone can hate: You can monitor your observability into your serverless-running-on-servers with Observerless.
With Observerless principles, your application is always up. If it isn’t, how would you know? “Blameless postmortems” can now devolve into shouted accusations of “peeking.”
When you observe Observerless practices, hiring becomes so much easier. How many other companies can tell their SREs that on-call is wonderful–they will sleep like infants!
The idea has precedent, even. In any distributed system, you’ve got three kinds of execution: at least once, at most once, or exactly once. I am of course kidding; there is no guaranteed “exactly once” execution–there’s only “who the hell knows?” And if you’re applying Observerless principles, certainly not you–that’s a job for the monitoring/observability philosophers.
This bold approach follows in the footsteps of Chaos Engineering, who turned “breaking things in production” from “the reason we fired Taylor” to “an incredibly lucrative career path.” With this bold methodology, you can unlock your inner lazy sack of crap.
Even vendors are jumping on board with the idea of helping you not observe anything that happens within your environment and you should absolutely want to do business with them: without their help, your employees are bound to screw it up and observe something.
While this may sound like it’s far off in the future and absolutely untenable for at least another five years, that’s really what being a thought leader is all about. How did I see this coming before the rest of you?
Simple: I peeked.
Corey is a Cloud Economist at the Quinn Advisory Group and an advisor to ReactiveOps. He has a history as an engineering director, public speaker, and cloud architect. Corey specializes in helping companies address horrifying AWS bills, and curates LastWeekinAWS.com, a weekly newsletter summarizing the latest in AWS news, blogs, and tips, sprinkled with snark.
Outside of his professional work, Corey is known for overdressing, telling entertaining stories, and carrying a cigarette case full of tiny umbrellas.