Manager after google went down: Hey I don't want our project to go like google. Once one server went down, gmail went down, then youtube, then drive, and everything else with it. Could we separate all of our features into different servers so that if one thing goes down, the rest aren't affected?

NO

It's the chicken and the egg, as consumers we get upset that there are no other options.

On the other hand we also don't seem to put much into exploring / supporting /using other options...

So now we're used to it and confused why equivalent / easy to use alternatives don't just appear for us.

😂

Just migrate all the fancy server stuff to p2p! How hard can it be?!

While you're at it, add some git as well, I heard it's great for syncing offline/online data!

@SuspiciousBug

npm i google-but-PHP

Just tell him you need a high availability zone. To create one, you have to split the building in half, and add productivity enhancing things like rock walls, massage chairs, and latte bars as the conduit between them. This creates a buffer that massages then irascible electrons so the system does fail. When the electrons are happy, they naturally walk over to the systems that work.

Yes. It's called a cluster. Another thing is called DR [either aa or ap], or multiple DRs.

But... Do you guys have separate applications running on the same servers?

Or is he just dumb and compared 2 features of a service to 2 entire applications?

>implying google has everything running on one computer

As if that was the heart of the problem for *google*... The lack of servers...

@Nanos

Yeah I used to work with disaster recovery solutions (well that's what we called then).

Anyone going into the 'duplicate data across servers ... geography... etc' and doesn't treat those plans with respect / ... kinda full time job learning how it works.... gonna have a bad time.

Companies that did it right did testing every month to see how fail overs work and check data and etc.