Hacker News new | past | comments | ask | show | jobs | submit login

When they say "grep", they appear to literally be running grep on four billion documents[1]. I'm assuming that is a significantly more expensive operation than their normal search algorithm as it won't use whatever search tree they've got set up.

1. http://blog.blekko.com/2011/09/13/get-ready-to-grep-the-web/




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: