Yes, but also decentralized design is not enough on its own!
It’s entirely possible that all of Lemmy could be a sea of independent instances, that are all hosted on AWS because it’s cheap or easy. Or if even just the big ones run on AWS, sure my instance could be up technically, but there’s no content. And in either case we’d be in the same situation despite the federated design.
We need to make sure, as a community, that we maintain a diversity of tech stacks between the instances too!
Also I think technically, nit-pickily, Lemmy is federated and not decentralized. Because actual true decentralization actually doesn’t care where any instance is hosted. I’m sorry, I know…
We cant have a decentralized internet as explain in the last 2 seasons of Silicon Valley, as Idiocracy has become true so will Silicon valley.
All hail Mike Judge our prophet
Yeah, what’s up with that actually ?
Anyone with the Knowledge care to explain ?
I or others can go into more detail, but I’m guessing you do not want a super in depth answer?
One of the major cloud providers (aka renting a chunk of a data center) has a outage in the us-east-1 region. Because of internal dependencies on us-east-1, when that AWS region (aka data center) has problems it impacts service’s across all AWS regions. To end users, suddenly web sites will act strange, crash, or just not work as elements of their backend are having problems. Due to the raw size of AWS, when something like this happens vast swaths of the web will break.
Thank you for this clarifying information. So AWS seems heavily centralized in particular or that’s just how it goes for those types of services ?
There’s only a few big web hosts really. It’s like with regular services like e-commerce or whatever. There’s Amazon, Oracle, and Microsoft. Over time they’ve either bought or undercut competitors until it’s one big centralized web world. There’s others but they’re small fish to those three.
What do you need someone to be Knowledgeable about? AWS US East 1 is down
This, precisely; a knowledge I was deprived of.
Why are we saying knowledge?
Are you sure you’re ready for that knowledge?
To extend what others have said. A main service in the most popular public clowd (AWS/Amazon Web Services) is down, which affects all customers that rely on that service, and they are many.
AWS does a lot of redundancy, but their identity management service (IAM) can’t do much redundancy without adding vulnerabilities to it. That service had an issue and nothing else can take the batton for security reasons.
Why companies do not have a contingency plan? Because they decided to fire some of their IT teams, giving the money to Amazon instead (you need to have an MBA to understand it, cocaine addiction or a combination).
I maybe wrong, my major was natural sciences but I chose not to be poor.
The computers that these services run on are not able to connect to the internet
These services all rent computers in the same building.
It’s stupid, I know.
sorry I’m a boy so i went to jupiter
The irony is just too much not to share.
