Dandandan commented on code in PR #16443:
URL: https://github.com/apache/datafusion/pull/16443#discussion_r2203255187
##########
datafusion/physical-plan/src/joins/utils.rs:
##########
@@ -843,24 +844,56 @@ pub(crate) fn apply_join_filter_to_indices(
probe_indices: UInt32Array,
filter: &JoinFilter,
build_side: JoinSide,
+ max_intermediate_size: Option<usize>,
) -> Result<(UInt64Array, UInt32Array)> {
if build_indices.is_empty() && probe_indices.is_empty() {
return Ok((build_indices, probe_indices));
};
- let intermediate_batch = build_batch_from_indices(
- filter.schema(),
- build_input_buffer,
- probe_batch,
- &build_indices,
- &probe_indices,
- filter.column_indices(),
- build_side,
- )?;
- let filter_result = filter
- .expression()
- .evaluate(&intermediate_batch)?
- .into_array(intermediate_batch.num_rows())?;
+ let filter_result = if let Some(max_size) = max_intermediate_size {
+ let mut filter_results =
+ Vec::with_capacity(build_indices.len().div_ceil(max_size));
+
+ for i in (0..build_indices.len()).step_by(max_size) {
+ let end = min(build_indices.len(), i + max_size);
+ let len = end - i;
+ let intermediate_batch = build_batch_from_indices(
+ filter.schema(),
+ build_input_buffer,
+ probe_batch,
+ &build_indices.slice(i, len),
+ &probe_indices.slice(i, len),
+ filter.column_indices(),
+ build_side,
+ )?;
+ let filter_result = filter
+ .expression()
+ .evaluate(&intermediate_batch)?
+ .into_array(intermediate_batch.num_rows())?;
+ filter_results.push(filter_result);
Review Comment:
I think batch coalescer won't make this faster as this is buffering
everything in memory anyway.
The main idea would be to apply the filters iteratively to the incoming
`RecordBatch` instead of the indices, so we have to change the API /
implementation a bit more.
It's fine to leave this as future change.
--
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: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
