Does anyone know if citusdb or enterprisedb improve on the postgresql issues mentioned in the post vs last postgresql version?
Why Uber Engineering Switched from Postgres to MySQL
31–40 of 306 posts
Re: Why Uber Engineering Switched from Postgres to MySQL
#32I've heard from technical leaders at multiple now well established unicorns how they'd never use postgres or switched from postgres simply because MySQL has a lot more tooling built it and many more people are exposed to its shortcomings at "web scale" so that it's very well known where and when things will break. Disclaimer, I'm a hardcore Postgres user myself, but I also keep tabs on the other tools.
I've been a hardcore and exclusive Postgres user since 2004. Always rolled my eyes when folks say they use MySQL or moved to it from Postgres. Reading about that bug sent shiver's down my spine. I know no system is bug free, but Postgres is one of those systems that have held up very well for me, and now I'm super paranoid. :-( I guess I understand tho, I moved from BSD to Linux for more tooling.
Re: Why Uber Engineering Switched from Postgres to MySQL
#33I like the sample data they have used: id first last birth_year 1 Blaise Pascal 1623 2 Gottfried Leibniz 1646 3 Emmy Noether 1882 4 Muhammad al-Khwārizmī 780 5 Alan Turing 1912 6 Srinivasa Ramanujan 1887 7 Ada Lovelace 1815 8 Henri Poincaré 1854
Mind explaining the significance? I didn't pick up on it.
Re: Why Uber Engineering Switched from Postgres to MySQL
#34I like the sample data they have used: id first last birth_year 1 Blaise Pascal 1623 2 Gottfried Leibniz 1646 3 Emmy Noether 1882 4 Muhammad al-Khwārizmī 780 5 Alan Turing 1912 6 Srinivasa Ramanujan 1887 7 Ada Lovelace 1815 8 Henri Poincaré 1854
Mind explaining the significance? I didn't pick up on it.
They're all famous mathematicians
Re: Why Uber Engineering Switched from Postgres to MySQL
#35The connection handling section was surprising to me, reading that Postgres uses a process per connection! This is pretty shocking to me, in a bad way.
pgbouncer works pretty well for that. I kinda wish it was part of the default Postgres install.
How does pgbouncer work with Amazon RDS (where you can't install separate software)?
Re: Why Uber Engineering Switched from Postgres to MySQL
#36I like the sample data they have used: id first last birth_year 1 Blaise Pascal 1623 2 Gottfried Leibniz 1646 3 Emmy Noether 1882 4 Muhammad al-Khwārizmī 780 5 Alan Turing 1912 6 Srinivasa Ramanujan 1887 7 Ada Lovelace 1815 8 Henri Poincaré 1854
Mind explaining the significance? I didn't pick up on it.
They're all famous scientists/mathematicians.
Re: Why Uber Engineering Switched from Postgres to MySQL
#37this reads like a laundry list of buzzwords that were designed to justify not throwing any effort into postgresql and just going with a new shiny toy (not mysql. yes. i know it's been around for a while). it happens everywhere.
From the post: > [...] This design difference means that the MySQL replication binary log is significantly more compact than the PostgreSQL WAL stream. Doesn't sound like what you described at all.
Isn't WAL-based replication more reliable?
Re: Why Uber Engineering Switched from Postgres to MySQL
#38I wonder what the design decisions are behind (or what it would take) to make Postgres store secondary indexes on disk like InnoDB does. Sure, the extra index lookup through the primary index is a cost, but it seems like write-amplification can sure be a greater concern too. Ultimately, it would be nice if Postgres gave the DBA a choice of — if not move outright to — secondary-index indirection through the primary index like InnoDB does.
Re: Why Uber Engineering Switched from Postgres to MySQL
#39I like the sample data they have used: id first last birth_year 1 Blaise Pascal 1623 2 Gottfried Leibniz 1646 3 Emmy Noether 1882 4 Muhammad al-Khwārizmī 780 5 Alan Turing 1912 6 Srinivasa Ramanujan 1887 7 Ada Lovelace 1815 8 Henri Poincaré 1854
Mind explaining the significance? I didn't pick up on it.
old mathematicians
Re: Why Uber Engineering Switched from Postgres to MySQL
#40This is a fantastic read. I hope the pg folks can turn as many of the issues brought up here into bug reports as possible (I think many of the issues, especially re: replication, are known), this kind of feedback is invaluable.