Timeline for Autodetect ‘random_page_cost’ vs ‘seq_page_cost’
Current License: CC BY-SA 3.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 5, 2022 at 14:12 | comment | added | Mikko Rantalainen |
@a_horse_with_no_name Most SSD devices have huge performance difference between sequential data vs true random access. The difference is not nearly as big as with HDDs but the difference is still there. Practically the only exception is Intel Optane SSD – those devices have so good random read performance that there you can probably set random_page_cost to seq_page_cost without any problem even in edge cases.
|
|
| S Apr 27, 2018 at 9:29 | history | bounty ended | guettli | ||
| S Apr 27, 2018 at 9:29 | history | notice removed | guettli | ||
| Apr 27, 2018 at 9:29 | vote | accept | guettli | ||
| Apr 25, 2018 at 1:16 | history | tweeted | twitter.com/StackDBAs/status/988949832135069697 | ||
| Apr 24, 2018 at 10:43 | history | edited | guettli | CC BY-SA 3.0 |
vs
|
| Apr 23, 2018 at 9:29 | history | edited | guettli | CC BY-SA 3.0 |
added 306 characters in body
|
| Apr 23, 2018 at 9:27 | comment | added | guettli | @a_horse_with_no_name I don't understand you comment. You say they are not used for writing. That's true. But why not create some tables do some benchmarking and finally show matching values? I updated the question. | |
| Apr 22, 2018 at 15:03 | comment | added | user1822 |
random_page_cost and seq_page_cost only apply for reading data. I don't think they are used for any write activity.
|
|
| Apr 22, 2018 at 5:31 | comment | added | Evan Carroll | @a_horse_with_no_name two reads from different chips can be pipelined with an SSD, and if you write sequentially I believe the writes get balanced to the different chips. | |
| Apr 22, 2018 at 5:26 | answer | added | Evan Carroll | timeline score: 13 | |
| Apr 20, 2018 at 9:37 | comment | added | user1822 |
For a SSD, I think it's safe to set random_page_cost to 1 as there is no difference in performance between random access and sequential access on a SSD if I'm not mistaken
|
|
| S Apr 20, 2018 at 9:30 | history | bounty started | guettli | ||
| S Apr 20, 2018 at 9:30 | history | notice added | guettli | Draw attention | |
| Apr 17, 2018 at 12:25 | history | asked | guettli | CC BY-SA 3.0 |