Skip to content

GH-3464: Improve DeltaByteArrayWriter.writeBytes to avoid unnecessary allocation and scalar prefix comparison #3465

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

Open
arouel wants to merge 2 commits into
base: master
Choose a base branch
from
Open

GH-3464: Improve DeltaByteArrayWriter.writeBytes to avoid unnecessary allocation and scalar prefix comparison #3465

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b7839f2
GH-3464 Improve `DeltaByteArrayWriter.writeBytes` to avoid unnecessar…
arouel Apr 6, 2026
60fe238
GH-3464 Add regression test
arouel Apr 6, 2026
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
Next Next commit
GH-3464 Improve DeltaByteArrayWriter.writeBytes to avoid unnecessar… 
…y allocation and scalar prefix comparison
  • Loading branch information
@arouel
arouel committed Apr 6, 2026
commit b7839f24d51101257e073216704aaadb740062d5
20 changes: 13 additions & 7 deletions ...umn/src/main/java/org/apache/parquet/column/values/deltastrings/DeltaByteArrayWriter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@
*/
package org.apache.parquet.column.values.deltastrings;

import java.util.Arrays;
import org.apache.parquet.bytes.ByteBufferAllocator;
import org.apache.parquet.bytes.BytesInput;
import org.apache.parquet.column.Encoding;
Expand Down Expand Up @@ -88,14 +89,19 @@ public String memUsageString(String prefix) {

@Override
public void writeBytes(Binary v) {
int i = 0;
byte[] vb = v.getBytes();
int length = previous.length < vb.length ? previous.length : vb.length;
// find the number of matching prefix bytes between this value and the previous one
for (i = 0; (i < length) && (previous[i] == vb[i]); i++)
;
byte[] vb = v.getBytesUnsafe();
int length = Math.min(previous.length, vb.length);
// Find the number of matching prefix bytes between this value and the previous one.
// Arrays.mismatch is intrinsified by the JVM to use SIMD instructions.
int i = Arrays.mismatch(previous, 0, length, vb, 0, length);
if (i < 0) {
i = length; // all bytes in the common range matched
}
prefixLengthWriter.writeInteger(i);
suffixWriter.writeBytes(v.slice(i, vb.length - i));
previous = vb;
// Retain an owned copy for prefix comparison with the next value.
// getBytesUnsafe() may return the backing array directly, so we must copy
// if the Binary's backing bytes may be reused by the caller.
previous = v.isBackingBytesReused() ? v.getBytes() : vb;
}
}