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
For every element of list on marshaling graphql object we get one goroutine. It's bad for performance on load tests. We should use goroutine for network call or another system call. Or hard parallel calculations. By marshaling isn't hard calculation.
What did you expect?
I propose to add "@Concurrent" directive for control the behavior. That lists consists concurrent type will have generated code as is now. But for lists with not concurrent type (without @Concurrent directive) will not have concurrent code.
The text was updated successfully, but these errors were encountered:
krupyansky
changed the title
Add @concurrent directive for types
[proposal] Add @concurrent directive for types
Jul 30, 2024
What happened?
For every element of list on marshaling graphql object we get one goroutine. It's bad for performance on load tests. We should use goroutine for network call or another system call. Or hard parallel calculations. By marshaling isn't hard calculation.
What did you expect?
I propose to add "@Concurrent" directive for control the behavior. That lists consists concurrent type will have generated code as is now. But for lists with not concurrent type (without @Concurrent directive) will not have concurrent code.
The text was updated successfully, but these errors were encountered: