Technology

Her real comment was that there are only 3 major cloud providers they can consider: AWS, GCP, and Azure. They chose AWS and AWS only. So there are a few options for them going forward — 1) keep doing what they’re doing and hope a single cloud provider can improve reliability, 2) modify their architecture to a multi-cloud architecture given the odds of more than one major provider going down simultaneously is much rarer, or 3) build their own datacenters/use colos which have a learning curve yet are still viable alternatives. Those that are serious about software own their own hardware, after all.

Each choice has its strengths and drawbacks. The economics are tough with any choice. Comes down to priorities, ability to differentiate, and value in differentiation :)

So much talking out of ass in these comments.

Federation/decentralization is great. It's why we're here on Lemmy.

It also means you expect everyone involved, people you've never met or vetted, to be competent and be able to shell out the cash and time to commit to a certain level of uptime. That's unacceptable for a high SLA product like Signal. Hell midwest.social, the Lemmy instance I'm on, is very often quite slow. I and others put up with it because we know it's run by one person on one server that he's presumably paying for himself. But that doesn't reflect Lemmy as a whole.

AWS isn't just a bunch of servers. They have dedicated services for database clusters, cache store, data warehouse, load balancing, container clusters, kubernetes clusters, CDN, web access firewall, to name just a few. Every region has multiple datacenters, the largest by far of which is North Virginia's. By default most people use one DC but multi region while being a huge expensive lift is something they already have tools to assist with. Also, and maybe most importantly, AWS, Azure and GCP run their own backbones between the datacenters rather than rely on the shared one that you, me, and most other smaller DCs are using.

I'm a DevOps Engineer but I'm no big tech fan. I run my own hobby server too. Amazon is an evil company. But the claim that "multi cloud is easy, smaller CSPs are just as good" is naive at best.

Ideally some legislation comes in and forces these companies to simplify the process for adopting multi cloud, because right now you have to build it all yourself and it becomes still very imperfect when you start to factor things like databases and DNS, and this is what they rely on hard for vendor lock-in.

Session is a decentralized alternative to signal. It doesn't require a phone number and all traffic is routed through a tor like onion network. Relays are run by the community and relay operators are rewarded with some crypto token for their troubles. To prevent bad actors from attacking the network, in order to run a relay you have to stake some of those tokens first and if your node misbehaves thay will get slashed.

SimpleX literally solves the messaging problem. You can bounce through their default relay nodes or run your own to use exclusively or add to the mix. It's all very transparent to end users.

At most, aws outage would have only affected chats relayed on those aws servers.

SimpleX also doesn't require a fukkin phone number.

They are serving 1on1 chats and group chats. That practically partitions itself. There are many server lease options all over the world. My assumption is that they use some AWS service and now can't migrate off. But you need an oncall team anyway so you aren't buying that much convenience.

Didn't only 1 AWS region go down? maybe before even thinking about anything else they should focus on redundancy within AWS

Tangent: Jami is p2p, so the only risk of going offline is if everyone in the groups go offline. It does lack several quality of life features, though.

Why is it that only the larger cloud providers are acceptable? What's wrong with one of the smaller providers like Linode/Akamai? There are a lot of crappy options, but also plenty of decent ones. If you build your infrastructure over a few different providers, you'll pay more upfront in engineering time, but you'll get a lot more flexibility.

For something like Signal, it should be pretty easy to build this type of redundancy since data storage is minimal and sending messages probably doesn't need to use that data storage.

The phrasing of the quotes is very "I sure hope someone comes along and fixes this for me because I'm not going to"