Hacker News Reader

The high end estimate would be for 25 million households and would mostly go to customer service [0]. It may well be too high, but "absolutely absurd" doesn't seem fair.

[0]: https://www.irs.gov/pub/irs-pdf/p5788.pdf#page=18

View on HN

The customer support costs make that seem more reasonable (or at least I'm not familiar enough to judge), but $7M in technology costs to go from 5M to 25M users without an increase in product scope seems very steep. Like even with 100KB of data stored per user, which seems like way more than necessary, that's only 2.5 TB. So storage isn't the issue. And if everyone did their returns in the last possible hour and did 100 requests/return, that's ~700k requests/second. In actuality the load should be far lower; maybe it'd peak in the thousands. I'm not seeing why there would be any variable technology cost at that level, much less $7M.

Putting some concrete numbers on the infrastructure, for $500k you could have like 10 servers each with 60 cores and 32 TB of mirrored NVMe storage. Any one server could handle the load while being almost completely idle (i.e. you're really just buying extreme redundancy). So what are the other $6.5M for, and how is that an annual variable tech budget?

Reply on news.ycombinator.com