Skip to content

Commit 09abf42

Browse files
praveen-influxpraveen-influx
committed
chore: more debug logs
1 parent 8598a99 commit 09abf42

File tree

2 files changed

+97
-40
lines changed

2 files changed

+97
-40
lines changed

‎influxdb3_write/src/write_buffer/mod.rs

+1-1
Original file line numberDiff line numberDiff line change
@@ -2051,7 +2051,7 @@ mod tests {
20512051
);
20522052
}
20532053

2054-
#[tokio::test]
2054+
#[test_log::test(tokio::test)]
20552055
async fn notifies_watchers_of_snapshot() {
20562056
let obj_store: Arc<dyn ObjectStore> = Arc::new(InMemory::new());
20572057
let (wbuf, _, _) = setup(

‎influxdb3_write/src/write_buffer/queryable_buffer.rs

+96-39
Original file line numberDiff line numberDiff line change
@@ -6,10 +6,10 @@ use crate::write_buffer::table_buffer::TableBuffer;
66
use crate::{ChunkFilter, ParquetFile, ParquetFileId, PersistedSnapshot};
77
use anyhow::Context;
88
use arrow::{
9-
array::AsArray,
9+
array::{AsArray, UInt64Array},
10+
compute::take,
1011
datatypes::TimestampNanosecondType,
1112
record_batch::RecordBatch,
12-
row::{RowConverter, SortField},
1313
};
1414
use async_trait::async_trait;
1515
use data_types::{
@@ -203,8 +203,7 @@ impl QueryableBuffer {
203203
let table_name =
204204
db_schema.table_id_to_name(table_id).expect("table exists");
205205
// mapping between time to main record batch array's index
206-
let mut smaller_chunks: HashMap<i64, (MinMax, Vec<usize>)> =
207-
HashMap::new();
206+
let mut smaller_chunks: HashMap<i64, (MinMax, Vec<u64>)> = HashMap::new();
208207
let smaller_duration = Duration::from_secs(10).as_nanos() as i64;
209208
let all_times = chunk
210209
.record_batch
@@ -214,59 +213,62 @@ impl QueryableBuffer {
214213
.values();
215214
for (idx, time) in all_times.iter().enumerate() {
216215
let smaller_chunk_time = time - (time % smaller_duration);
217-
let (min_max, vec_indices) =
218-
smaller_chunks.entry(smaller_chunk_time).or_insert_with(|| {
219-
(MinMax::new(i64::MAX, i64::MIN), Vec::new())
220-
});
216+
let (min_max, vec_indices) = smaller_chunks
217+
.entry(smaller_chunk_time)
218+
.or_insert_with(|| (MinMax::new(i64::MAX, i64::MIN), Vec::new()));
221219

222220
min_max.update(*time);
223-
vec_indices.push(idx);
221+
vec_indices.push(idx as u64);
222+
}
223+
224+
let total_row_count = chunk.record_batch.column(0).len();
225+
226+
for (smaller_chunk_time, (min_max, all_indexes)) in smaller_chunks.iter() {
227+
debug!(
228+
?smaller_chunk_time,
229+
?min_max,
230+
num_indexes = ?all_indexes.len(),
231+
?total_row_count,
232+
">>> number of small chunks");
224233
}
225234

226235
// at this point we have a bucket for each 10 sec block, we can create
227236
// smaller record batches here but maybe wasteful if we ever needed one
228237
// batch (let's see how this works first and then decide what can happen)
229238
let batch_schema = chunk.record_batch.schema();
239+
debug!(schema = ?chunk.schema, ">>> influx schema");
240+
debug!(arrow_schema = ?batch_schema, ">>> batch schema");
230241
let parent_cols = chunk.record_batch.columns();
231-
let fields = batch_schema
232-
.fields()
233-
.iter()
234-
.map(|field| SortField::new(field.data_type().clone()))
235-
.collect();
236-
debug!(?fields, ">>> schema fields");
237-
238-
let converter =
239-
RowConverter::new(fields).expect("row converter created from fields");
240-
let rows = converter
241-
.convert_columns(parent_cols)
242-
.expect("convert cols to rows to succeed");
243-
244-
for (smaller_chunk_time, (min_max, all_indexes)) in smaller_chunks.iter() {
245-
// create a record batch using just all_indexes from parent recordbatch
246-
let all_rows = all_indexes
247-
.iter()
248-
.map(|idx| rows.row(*idx))
249-
.collect::<Vec<_>>();
250-
251-
let child_cols = converter
252-
.convert_rows(all_rows)
253-
.expect("should convert rows back to cols");
254242

243+
for (smaller_chunk_time, (min_max, all_indexes)) in
244+
smaller_chunks.into_iter()
245+
{
246+
let mut smaller_chunk_cols = vec![];
247+
let indices = UInt64Array::from_iter(all_indexes);
248+
for arr in parent_cols {
249+
let filtered =
250+
take(&arr, &indices, None)
251+
.expect("index should be accessible in parent cols");
252+
253+
debug!(smaller_chunk_len = ?filtered.len(), ">>> filtered size");
254+
smaller_chunk_cols.push(filtered);
255+
}
256+
debug!(smaller_chunk_len = ?smaller_chunk_cols.len(), ">>> smaller chunks size");
255257
let smaller_rec_batch =
256-
RecordBatch::try_new(Arc::clone(&batch_schema), child_cols)
258+
RecordBatch::try_new(Arc::clone(&batch_schema), smaller_chunk_cols)
257259
.expect("create smaller record batch");
258260
let persist_job = PersistJob {
259261
database_id: *database_id,
260262
table_id: *table_id,
261263
table_name: Arc::clone(&table_name),
262-
chunk_time: *smaller_chunk_time,
264+
chunk_time: smaller_chunk_time,
263265
path: ParquetFilePath::new(
264266
self.persister.node_identifier_prefix(),
265267
db_schema.name.as_ref(),
266268
database_id.as_u32(),
267269
table_name.as_ref(),
268270
table_id.as_u32(),
269-
*smaller_chunk_time,
271+
smaller_chunk_time,
270272
snapshot_details.last_wal_sequence_number,
271273
),
272274
batch: smaller_rec_batch,
@@ -277,6 +279,63 @@ impl QueryableBuffer {
277279
};
278280
persisting_chunks.push(persist_job);
279281
}
282+
// let fields = batch_schema
283+
// .fields()
284+
// .iter()
285+
// .map(|field| SortField::new(field.data_type().clone()))
286+
// .collect();
287+
// debug!(?fields, ">>> schema fields");
288+
//
289+
// let converter =
290+
// RowConverter::new(fields).expect("row converter created from fields");
291+
// debug!(?converter, ">>> converter");
292+
//
293+
// let rows = converter
294+
// .convert_columns(parent_cols)
295+
// .expect("convert cols to rows to succeed");
296+
// debug!(?rows, ">>> all rows");
297+
//
298+
// for (smaller_chunk_time, (min_max, all_indexes)) in smaller_chunks.iter() {
299+
//
300+
// // create a record batch using just all_indexes from parent recordbatch
301+
// let all_rows = all_indexes
302+
// .iter()
303+
// .map(|idx| rows.row(*idx))
304+
// .collect::<Vec<_>>();
305+
// debug!(?rows, ">>> all filtered child rows");
306+
//
307+
// // hmmm this conversion turns Dictionary types to StringArray, not sure
308+
// // why
309+
// let child_cols = converter
310+
// .convert_rows(all_rows)
311+
// .expect("should convert rows back to cols");
312+
// debug!(?child_cols, ">>> all child cols");
313+
//
314+
// let smaller_rec_batch =
315+
// RecordBatch::try_new(Arc::clone(&batch_schema), child_cols)
316+
// .expect("create smaller record batch");
317+
// let persist_job = PersistJob {
318+
// database_id: *database_id,
319+
// table_id: *table_id,
320+
// table_name: Arc::clone(&table_name),
321+
// chunk_time: *smaller_chunk_time,
322+
// path: ParquetFilePath::new(
323+
// self.persister.node_identifier_prefix(),
324+
// db_schema.name.as_ref(),
325+
// database_id.as_u32(),
326+
// table_name.as_ref(),
327+
// table_id.as_u32(),
328+
// *smaller_chunk_time,
329+
// snapshot_details.last_wal_sequence_number,
330+
// ),
331+
// batch: smaller_rec_batch,
332+
// // this schema.clone() can be avoided?
333+
// schema: chunk.schema.clone(),
334+
// timestamp_min_max: min_max.to_ts_min_max(),
335+
// sort_key: table_buffer.sort_key.clone(),
336+
// };
337+
// persisting_chunks.push(persist_job);
338+
// }
280339
}
281340
}
282341
}
@@ -494,6 +553,7 @@ impl QueryableBuffer {
494553
}
495554
}
496555

556+
#[derive(Debug)]
497557
struct MinMax {
498558
min: i64,
499559
max: i64,
@@ -505,10 +565,7 @@ impl MinMax {
505565
// it's good to start with i64::MAX for min and i64::MIN
506566
// for max in loops so this type unlike TimestampMinMax
507567
// doesn't check this pre-condition
508-
Self {
509-
min,
510-
max
511-
}
568+
Self { min, max }
512569
}
513570

514571
fn update(&mut self, other: i64) {

0 commit comments

Comments
 (0)