Samyak2 commented on code in PR #20500:
URL: https://github.com/apache/datafusion/pull/20500#discussion_r3130904791
##########
datafusion/physical-plan/src/common.rs:
##########
@@ -204,6 +205,31 @@ pub fn can_project(
}
}
+/// Return a new `RecordBatch` with [`StringViewArray::gc`] called on such
columns (if any).
+pub(crate) fn gc_stringview_arrays(
+ batch: RecordBatch,
+) -> Result<RecordBatch, ArrowError> {
+ let mut new_columns: Vec<Arc<dyn Array>> =
Vec::with_capacity(batch.num_columns());
+
+ let mut arr_mutated = false;
+ for array in batch.columns() {
+ if let Some(string_view_array) =
array.as_any().downcast_ref::<StringViewArray>()
+ {
+ let new_array = string_view_array.gc();
Review Comment:
Alright, let me try this.
But if your guess is correct, then we're blocking the next IO call from
happening due to CPU compute above the data source. So keeping the coalesce
downstream is just removing some compute from that path. The core issue seems
to be that we're not doing pre-fetching of IO while the compute is running?
I'm not sure, but I hope to get more insights when I try this out.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
