Microsoft moved late with Azure, but as they had so many customers in B2B already with SQL Server and Office, that they soon grew market share. Google Cloud followed and acquired companies like Firebase to differentiate.
Market Share 2026
AWS 28%
Azure 21%
Google 14%
https://www.statista.com/chart/18819/worldwide-market-share-... "The systems are down, it's a global Microsoft Cloud issue affecting everyone"
vs
"The systems are down, Bob from my team made an error but we are working to fix it"We're small (< 25 people). We maintained a single rack in a colo, and all the servers were showing their age. We weighed lease cost for replacement (along with hassle of actually doing it), and migrated into AWS last January.
It's been ... messy. I'm not certain we'll stay. It's easier in the sense that we can bring up arbitrarily many new machines without needing to call Dell when a host gets full, but it's definitely not cheaper. And weirdball AWS-only things get in the way. Oh, and literally nothing is simple or easy in administration or setup.
If I had it to do over again, I think I'd have bitten the bullet and leased more boxes instead.
Move to AWS? All techies want to add that to their CV so sure why not... Who cares if it's the best decision as long as the CVs get improved.
============================
[1] My personal bugbear are techies who are adamant that nothing but google chrome can be used by the regardless of the fact that when pressed they can't find a single site that doesn't work on FF. Usually the goalposts are moved at some point in the discussion from "sites don't work on FF" to performance even though FF with uBlock origin cuts actual downloading of ads to zero compared to the uBlock lite on chrome. They'll next complain about memory used in FF compared to google regardless of the lack of any measurements and even though these same techies are using bloated electron apps daily anyway!
As you can tell, this really is a personal bugbear of mine.
AWS was also cheaper in the beginning but has been slowly ratcheting up the pricing making colocation or on-premise viable again for some workloads which have steady compute demands.
For the first (cloud vs. non-cloud) it's clear that cloud has some advantages because you don't have to manage hardware.
For the second (AWS vs. other-cloud), at least in my experience, AWS is simply the best experience available. If anything the other cloud providers go out of their way to offer special deals and cozy customer relationships to try to get customers to switch, but AWS is still the gold standard in availability and uptime and ease of use.
The gap is narrowing as the field becomes more commoditized though. And I think that's what we're really seeing here. As things like k8s get more common, solutions are becoming more and more substrate-independent, so companies hunt for smaller, more local, cheaper, ambitious alternatives. Everyone is scared of vendor lock-in so it's getting harder for cloud providers to offer distinguishing features.
We all just forgot you can rent a server huh
I really like AWS. But I feel like there is a lot of money to be made in consultants overselling services built on it. And a lot of traditionally expensive IT teams are being replaced with even more expensive cloud "architects".
Then it got not only complicated but complex, and full of vendor lock in.
It used to be easy to orchestrate, there were terraform modules for everything. I chose it multiple times for that exact reason.
The cost structure used to also be good, you knew how much your bill was gonna be, you could get cheaper pricing if you went for spot instances etc.
It's expensive, but you get what you pay for.
They’re not idiots. A lot of thought went into it, and the idea it’s all just incompetent BS dreamt up on a golf course is madness.
and vertical scale
and fuck fuck fuck load of features.
I agree if we are operating a cat meme distribution business that AWS starts to look a bit overkill. Something like a banking consultancy is a different matter. Good luck passing due diligence as a bank's vendor when you are also physically responsible for the machines. You can do this, but it sucks and it's your entire job + overtime + on call.
AWS has a dedicated feature (artifact) that can serve as the foundation of your compliance package. If you are racking your own machines you get to start with a blank pile of forms.
This is not comparable to the situation in a company like Airbus, with perhaps 10k people of every imaginable expertise employed in just the IT department. Not comparable to most companies that are 1/50th of Airbus size either.
For a largish company (1000+ people, 500M+ turnover), assigning ten FTE:s to managing the racks and general core infra on a 365/24 basis and another five lawyers and general bureaucrats to deal with the general paperwork is nothing.
The reason for the big cloud shift in the 2010s was other things. A lot of hype/lies about how fast, cheap and easy things would be. And in many companies probably also a feeling that internal IT were a bunch of morons who failed at anything you asked them to do, and that anything else must be better.
I think it's the other way around. At the extreme scales you don't want AWS.
If you're running a cat meme distribution business, a single senior skilled techie will cheaply get you set up on a VPS (or a bunch of them) with the services you need.
If you're a bank, you have all the budget in-house to babysit your own VPS and stand up your own equivalents to whatever the fuck it is you were using on AWS, and follow all regulations. Because, face it, as a bank, the type of regulations you meet by using AWS is a tiny fraction of all the regulations you have to follow, so there is literally a team within the bank tasked purely with making sure all banking regulations are followed, and that team exists even if you're on AWS for everything.
The only place AWS makes sense is if you're a company with no technical department in the first place.
The same people who were logging into AWS and managing stuff there.