Skip to content

LUCENE-10127: Minor speedup to doc values writes. #325

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.

Sign up for GitHub

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

Merged
jpountz merged 2 commits into from
Sep 28, 2021
Merged

LUCENE-10127: Minor speedup to doc values writes. #325

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6edcd28
LUCENE-10127: Minor speedup to doc values writes.
jpountz Sep 28, 2021
c30f783
Avoid Arrays#fill on large arrays.
jpountz Sep 28, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 18 additions & 3 deletions lucene/core/src/java/org/apache/lucene/codecs/lucene90/Lucene90DocValuesConsumer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -164,6 +164,13 @@ void update(long v) {
++numValues;
}

/** Accumulate state from another tracker. */
void update(MinMaxTracker other) {
min = Math.min(min, other.min);
max = Math.max(max, other.max);
numValues += other.numValues;
}

/** Update the required space. */
void finish() {
if (max > min) {
Expand All @@ -181,6 +188,13 @@ void nextBlock() {
private long[] writeValues(FieldInfo field, DocValuesProducer valuesProducer, boolean ords)
throws IOException {
SortedNumericDocValues values = valuesProducer.getSortedNumeric(field);
final long firstValue;
if (values.nextDoc() != DocIdSetIterator.NO_MORE_DOCS) {
firstValue = values.nextValue();
} else {
firstValue = 0L;
}
values = valuesProducer.getSortedNumeric(field);
int numDocsWithValue = 0;
MinMaxTracker minMax = new MinMaxTracker();
MinMaxTracker blockMinMax = new MinMaxTracker();
Expand All @@ -196,14 +210,14 @@ private long[] writeValues(FieldInfo field, DocValuesProducer valuesProducer, bo
// wrong results. Since these extreme values are unlikely, we just discard
// GCD computation for them
gcd = 1;
} else if (minMax.numValues != 0) { // minValue needs to be set first
gcd = MathUtil.gcd(gcd, v - minMax.min);
} else {
gcd = MathUtil.gcd(gcd, v - firstValue);
}
}

minMax.update(v);
blockMinMax.update(v);
if (blockMinMax.numValues == NUMERIC_BLOCK_SIZE) {
minMax.update(blockMinMax);
blockMinMax.nextBlock();
}

Expand All @@ -215,6 +229,7 @@ private long[] writeValues(FieldInfo field, DocValuesProducer valuesProducer, bo
numDocsWithValue++;
}

minMax.update(blockMinMax);
minMax.finish();
blockMinMax.finish();

Expand Down
6 changes: 5 additions & 1 deletion lucene/core/src/java/org/apache/lucene/util/packed/DirectWriter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,11 +89,15 @@ public void add(long l) throws IOException {
}

private void flush() throws IOException {
if (off == 0) {
return;
}
// Avoid writing bits from values that are outside of the range we need to encode
Arrays.fill(nextValues, off, nextValues.length, 0L);
encode(nextValues, 0, nextBlocks, 0, iterations);
final int blockCount =
(int) PackedInts.Format.PACKED.byteCount(PackedInts.VERSION_CURRENT, off, bitsPerValue);
output.writeBytes(nextBlocks, blockCount);
Arrays.fill(nextValues, 0L);
off = 0;
}

Expand Down