Timeline for PostgreSQL - Working with array of thousands of elements
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 6, 2018 at 19:27 | comment | added | Jukurrpa | Each item belongs to multiple groups (usually 15-20 of them), so I tried storing them as an int array in mytable but couldn't figure out how to index this properly. I updated the question with all the details. | |
| Apr 5, 2018 at 18:36 | comment | added | Evan Carroll | No, I mean. If you have 15,000 ids normally you're storing something in the ID, like whether or not the item is a kitchen-product, and rather than storing the group_id that corresponds to "kitchen product", you're trying to find all kitchen products by their ids. (which is bad for every reason) What is that those 15,000 ids represent? Why isn't it stored on the row itself? | |
| Apr 5, 2018 at 18:33 | comment | added | Jukurrpa | Just tried using the temporary table and it's slower, at least in the 15,000 ids case. As for creating groups in the schema itself do you mean a table with the ids I pass as an argument? I tried something like this but the performance was similar or worse than my current approach. I'll update the question with more details | |
| Apr 4, 2018 at 18:36 | history | answered | Evan Carroll | CC BY-SA 3.0 |