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

All of this looks accurate, but it's worth contextualizing: this is an optimization that bears the most fruit in "local frames of reference" -- the timescales where linear scans beat index lookups are likely to be strictly dominated by the network latency to transceive from the database. The conclusion is then that the optimization ~only optimizes for cases that effectively don't matter.


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

Search: