News for hackers. Stuff that matters.

A comment on Index bloat reduced in PostgreSQL v14. Parent comment

Re:Index bloat reduced in PostgreSQL v14 by smt88 on Friday September 10, 2021 @11:57PM (#28489320)
My DB knowledge is too patchy to give a good summary, but it was extensively discussed on HN several times[1].

In general it seems like they could have solved all of their problems within Postgres, but someone at Uber seemed to just know MySQL better and want to switch. They also made other major changes (like removing schemas from certain data stores) that likely gave them more of an advantage than switching could have.

1. https://news.ycombinator.com/item?id=12166585

[ Parent | on HN ]
Re:Index bloat reduced in PostgreSQL v14 by eloff on Saturday September 11, 2021 @01:56AM (#28489720)
I remember the article and the discussion, but the fundamental issue seemed to be the difference between MySQL and Postgres indices and the different trade-offs made. Postgres indices are better for read heavy workloads, and MySQL for write heavy. I don't see how this is representative of them not knowing what they are doing. Frankly it's condescending to think you understand their engineering problems better than they do.
[ Parent | on HN ]