4 ms·
For one, with a 1 file per row (entity) design, any larger query would require reading from potentially millions of files, which I imagine would be nightmarishl
by badpun 3y ago
For one, with a 1 file per row (entity) design, any larger query would require reading from potentially millions of files, which I imagine would be nightmarishly inefficient.
- xyheme 3y agoI think, unless column-oriented database is used, "one file per row" is similar to normal row-oriented database. Maybe there are some benchmarks in the past.