The fertile moment between completing the code, and running the code.
-
Show this thread
-
The third rewrite is going to be the one that lands, I can tell.
1 reply 0 retweets 8 likesShow this thread -
After three attempts at grand unified re-writes, I realized I could get 90% of the wins with a very small patch. I see 25x speed-ups in my classic point-in-polygon workload (countries vs populated places)https://github.com/postgis/postgis/pull/552 …
3 replies 1 retweet 13 likesShow this thread -
Replying to @pwramsey
Don't think that's safe across transactions. If the cache lives that long, a specific toastid could have been deleted and reused since you cached it.
1 reply 0 retweets 0 likes -
Replying to @AndresFreundTec
It only lives for the life of the statement. According to his Tom-ness, that was OK.
1 reply 0 retweets 0 likes -
Replying to @pwramsey
Good ;). I didn't see the single-statement-ness when skimming the diff for a few seconds ;)
1 reply 0 retweets 0 likes -
Replying to @AndresFreundTec
This seems pretty generic, and it's a win in cases where a nested loop join is being driven be a table with large objects. Worth bringing to core?
1 reply 0 retweets 0 likes
I'd like to have a solution to the problem in core, I'm not sure how far this gets us though. But we should investigate/discuss more.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.