2pb is nothing if we’re talking about a small datacenter backed up by the government. That said, Russia has a history of special kind of dumbassery

I can pretty much guarantee that the cost of creating an offsite backup is trivial compared to the budget used to collect and analyze those data. I can’t read Russian anymore and it’s probably not published in a discoverable way, but I’m going to offer up the possibility that the sat network, research scientist teams, sys admins, and everything else that goes into the portion of the Russian government’s budget for this work wouldn’t have even seen that as a rounding error. I’ve worked with US government budgets and I know how tight fisted committees can be, and while the USG isn’t Google in terms of writing checks for tech, and while the Russians are probably an order of magnitude or two poorer than our budgets, it’s still be a no brainer in terms of costs. Either they just didn’t think of it (which I’ve seen far more times than I can tell you about) or it got eliminated as a line item by some bureaucrats who don’t understand cost/benefit analysis (which we’ve all also seen), it wasn’t truly a cost thing. Compared to the price associated with sat launches and data analysis, $10-$20k/ month for data retention is nothing.

Also, I sort of suspect that these were dual use systems. When you’re talking about the sensing tech they’re using, there are the very obvious and direct intel applications.

Would they maintain their own private off-site backup or would they be in a cluster with other government agencies or renting out from a commercial operation?

The cost would be massive for you or I to utilise such services, less so for an agency, and it certainly isn’t difficult.

I understand science is generally always under funded and there’s probably some oligarch skimming off of their budget, but I still don’t see this being the win they think it is in any form. I can only hope the climate data is not lost to all time.