4 ms·

is it true that MySQL scales much better than postgres? for example ServiceNow is using customized mysql engine to store and search all data for all its clients

by slt2021 6y ago

is it true that MySQL scales much better than postgres? for example ServiceNow is using customized mysql engine to store and search all data for all its clients. Doing something like that in pg is probably extremely hard. Would love to hear people's thoughts on this

skunkworker 6y ago
This depends what you mean by "scaling". Postgres can easily handle 10s of Terabytes on a single server (though this is dependent upon your use case), but when it comes to multiple host sharded databases at the moment MySQL has some more mature options like Vitess [1], where your options in Postgres are things like Citus [2] which just recently open sourced their shard mover, but still have the HA and replication feature behind expensive enterprise pricing. For most people not dealing with the multiple terabyte database size, Postgres is what I would definitely recommend. But once you get above that, then there are no easy answers and you would need to look at what YouTube did with Vitess [3] [1] https://vitess.io/ https://vitess.io/ [2] https://www.citusdata.com/ https://www.citusdata.com/ [3] https://vitess.io/docs/overview/history/ https://vitess.io/docs/overview/history/
- slt2021 6y ago
  Thank you very much!
redis_mlc 6y ago
> is it true that MySQL scales much better than postgres? It's a strange question since "scales" depends on your site, but the short answer is yes: 1) Most of the FAANG companies run on MySQL, and they're approaching a trillion dollars in value each. 2) MySQL doesn't need a vacuum process, and isn't sensitive to number of indexes per table (write amplification) like Postgres is. 3) MySQL has more built-in replication methods than Postgres, which is very important for enterprises. 4) Postgres has a complex grant syntax that most devs don't understand. 5) Postgres has compliance certification issues around COPY FROM/COPY TO. There needs to be a grant to disable that per user. Both MySQL and Postgres are great Open Source databases. I use both, but prefer MySQL for the above reasons. I consider Postgres to be substantially weaker for enterprise use cases though. Source: DBA.

skunkworker 6y ago

This depends what you mean by "scaling". Postgres can easily handle 10s of Terabytes on a single server (though this is dependent upon your use case), but when it comes to multiple host sharded databases at the moment MySQL has some more mature options like Vitess [1], where your options in Postgres are things like Citus [2] which just recently open sourced their shard mover, but still have the HA and replication feature behind expensive enterprise pricing. For most people not dealing with the multiple terabyte database size, Postgres is what I would definitely recommend. But once you get above that, then there are no easy answers and you would need to look at what YouTube did with Vitess [3] [1] https://vitess.io/ https://vitess.io/ [2] https://www.citusdata.com/ https://www.citusdata.com/ [3] https://vitess.io/docs/overview/history/ https://vitess.io/docs/overview/history/

slt2021 6y ago
Thank you very much!

redis_mlc 6y ago

> is it true that MySQL scales much better than postgres? It's a strange question since "scales" depends on your site, but the short answer is yes: 1) Most of the FAANG companies run on MySQL, and they're approaching a trillion dollars in value each. 2) MySQL doesn't need a vacuum process, and isn't sensitive to number of indexes per table (write amplification) like Postgres is. 3) MySQL has more built-in replication methods than Postgres, which is very important for enterprises. 4) Postgres has a complex grant syntax that most devs don't understand. 5) Postgres has compliance certification issues around COPY FROM/COPY TO. There needs to be a grant to disable that per user. Both MySQL and Postgres are great Open Source databases. I use both, but prefer MySQL for the above reasons. I consider Postgres to be substantially weaker for enterprise use cases though. Source: DBA.