Timeline for PostgreSQL SSD vs HDD - why is there no difference in insert performance?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 10 at 23:12 | comment | added | Volksman | If you're running it on an enterprise level server (e.g. HP Proliant or similar) then there's a good chance that that writes to the HDDs are extremely fast because they're actually being written to a non volatile write cache. Ironic because writes to SSDs are much slower than reads so SSDs typically have their own RAM based write cache. | |
| Apr 23, 2021 at 14:02 | history | edited | jjanes | CC BY-SA 4.0 |
added 512 characters in body
|
| Apr 23, 2021 at 6:52 | comment | added | Leon Powałka | I read some articles about using SSD tablespaces for heavy write loads, so I assumed there would be some kind of a difference? Maybe I need to test it on bigger data? Or does it really not matter if I use SSD or HDD? | |
| Apr 22, 2021 at 14:57 | history | answered | jjanes | CC BY-SA 4.0 |