You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I am currently running a database with ~60000 STAC Collections.
By default, partitioning in done by "collection", which has worked well in the past but is causing problems in this instance.
I quickly run way too many partitions and queries get super slow, pretty fast.
I wanted to edit partitioning, for a datetime-based approach, for example having a partition per week.
However, it seems like it's not working because pgstac still tries to create _items_XXXXX partitions when loading items into the DB.
So my question is : how can I update the database parttioning to match my needs ?
Thanks :)
The text was updated successfully, but these errors were encountered:
@bjgiraudon Unfortunately, right now, hard coded partitioning by collections is very deeply ingrained in how pgstac works and the partition automation scripts. I do not see any reasonable way that we could remove that limitation without a lot of work. In order to have fully managed partition experience, we make the concession of losing some flexibility.
Do you know how I could improve database transactions? My implementation is quite heavy on writing, not so much on reading. I have 60k+ collections. Right now, my workflows are stuck on write operations, presumably looking for partitions.
Hi, I am currently running a database with ~60000 STAC Collections.
By default, partitioning in done by "collection", which has worked well in the past but is causing problems in this instance.
I quickly run way too many partitions and queries get super slow, pretty fast.
I wanted to edit partitioning, for a datetime-based approach, for example having a partition per week.
However, it seems like it's not working because pgstac still tries to create
_items_XXXXX
partitions when loading items into the DB.So my question is : how can I update the database parttioning to match my needs ?
Thanks :)
The text was updated successfully, but these errors were encountered: