It definitely can. Reasonable timeouts and redundant systems.
It's amazing the length some people are willing to go to to defend AWS marketing slogans as a source of truth. I've seen vendor lock-in before, but AWS seems to be unique in that people actually enjoy working with a vendor whose services go down randomly to the point where they blame themselves for not being "fault-tolerant".
Guess what, if your service is not required to be up because the consuming service is super tolerant to it timing out after 140 seconds, self-hosting it becomes even more of a no-brainer. After all, you clearly need none of the redundancy AWS features.
If it makes you feel better, everything I'm saying about AWS can be said about GCP as well.
Sorry, but AWS/GCP is infinitely better at managing infrastructure than you or your company will ever be.
What deathanatos wrote sounds awfully like problems with the service to me.
I don’t think S3 taking 100+ seconds to respond to a GET request can be solved by orchestration alone.