3 ms·

I have been burned before when a query on a newer db version was "optimized" in a different way that caused performance to drop but I have not yet had a query o

I have been burned before when a query on a newer db version was "optimized" in a different way that caused performance to drop but I have not yet had a query on the same version vary so drastically as to cause a problem. Is this a more common occurrence that I just have not encountered before?

williamdclt 3y ago
Not common (if it was anything more than exceptional, PG wouldn’t be a production ready product) but it happens: https://www.postgresql.org/message-id/17540-7aa1855ad5ec18b4%40postgresql.org https://www.postgresql.org/message-id/17540-7aa1855ad5ec18b4...
jandrewrogers 3y ago
It is fairly common in e.g. multi-terabyte Postgres databases. Beyond a certain scale, it can become impossible for the statistics collector to build an accurate model for large tables. When this happens, you can end up in the strange place where the query plans change every time you run analyze even though the data model and data distribution hasn't changed. I've seen operational issues at multiple companies due to the query plans randomly changing like this on large tables in Postgres.