-
Notifications
You must be signed in to change notification settings - Fork 679
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
the same query in the worker node is 100x fater than in coordinator node #7772
Comments
All we can do is guess unless you share the explain output. My guess is that Citus pulls all the data from the workers and does the filtering on the coordinator. Although I can not see why that would happen, the execution times suggest that. Check individual queries that were run on shards, and try to understand what is going on using the explain output. |
explain code
coordinator output
worker1 output
worker2 output
|
@hanefi ,I have update the explain output , need your help! |
I have setup citus(v12.1.15) cluster using postgreSQL16.4 with one coordinator and tow worker in 3 physical machine,with node run in docker. coordinator has 27TB size,worker1 has 22TB,worker2 has 6TB,while excute spatial query,in any of the worker node will get result in 400ms and 185 rows ,but excute in the coordinator node will cost 40s.
I have using bellow code to explain costs.
on coordinator the explain result like those:
all task planing and execution time in 0.x ms, but all task will took 40s
The text was updated successfully, but these errors were encountered: