Can this be reformulated: you store 10 trln objects (e.g. 100TB of 10 byte records), you lose 1 record each year.
Also curious what are the stats from other providers.
Roughly, yes.
> Also curious what are the stats from other providers.
As to other providers, most are 6+ 9s that I’ve looked at, with many in the 8-9 range. Anything over 8 is (as they admitted) essentially marketing porn and not a useful metric (for reasons they mentioned as well as ones said by other comments here).
I think a much bigger scandal is that all major laptop Operating System vendors (Microsoft and Apple) absolutely know when your laptop drive loses files or even is starting to go bad in some cases, and they NEVER tell the customer. I think an excellent product offering would be a 3rd party piece of software and cloud service which was a "verification service". It wouldn't store your files offsite, it would store the name, size, and SHA1 offsite and periodically check that no bits have been flipped on your local drive unless you intended it. For example, a week after I take a photo, I absolutely never want the photo to change. Ever. Same with music I (legally) download.
locally redundant storage: 99.999999999 % (11 9's)
zone redundant storage: 99.9999999999 % (12 9's)
geographically redundant storage: 99.99999999999999 % (16 9's)
Chunks of your data are going to be stored together, so it's a very small chance of losing a big block of 10 byte files. There's no failure mode that loses just one, and does so often.
It probably depends on their infrastructure, e.g. if storage is something like cassandra, records would be evenly distributed by key hash.
I agree that numbers likely are not like that though, just wanted to demonstrate that such calculation approach can bring unexpected conclusions.