Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I like it in theory but the IA hosts over 175PB of data. Wonder how many other producers could replicate that data.


Suppose you don't have ten hosts that each have 175PB of data but rather a million hosts that each have an average of 1.75TB, and therefore the equivalent of 10 full copies. And then something that periodically checks if there is any given subset of the data with too few copies and makes more.


I don’t have hard data to back this up, but I estimate that plenty of main Usenet binary providers easily exceed that.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: