Skip to content

CNDB-15423: Make TopKProcessor lazily load queryVector #2007

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
michaeljmarshall merged 1 commit into from
Sep 22, 2025
Merged

CNDB-15423: Make TopKProcessor lazily load queryVector #2007

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 4 additions & 5 deletions src/java/org/apache/cassandra/index/sai/plan/TopKProcessor.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,8 +93,9 @@ public class TopKProcessor
private final ReadCommand command;
private final IndexContext indexContext;
private final RowFilter.Expression expression;
private final VectorFloat<?> queryVector;
private final ColumnMetadata scoreColumn;
// Lazily compute this value, if needed.
private VectorFloat<?> queryVector = null;

private final int limit;

Expand All @@ -109,10 +110,6 @@ public TopKProcessor(ReadCommand command)

this.indexContext = indexAndExpression.left;
this.expression = indexAndExpression.right;
if (expression.operator() == Operator.ANN && !(Ordering.Ann.useSyntheticScore() && expression.column().isRegular()))
this.queryVector = vts.createFloatVector(TypeUtil.decomposeVector(indexContext, expression.getIndexValue().duplicate()));
else
this.queryVector = null;
this.limit = command.limits().count();
this.scoreColumn = ColumnMetadata.syntheticScoreColumn(expression.column(), FloatType.instance);
}
Expand Down Expand Up @@ -329,6 +326,8 @@ private float getScoreForRow(DecoratedKey key, Row row)
ByteBuffer value = indexContext.getValueOf(key, row, FBUtilities.nowInSeconds());
if (value != null)
{
if (queryVector == null)
queryVector = vts.createFloatVector(TypeUtil.decomposeVector(indexContext, expression.getIndexValue().duplicate()));
var vector = vts.createFloatVector(TypeUtil.decomposeVector(indexContext, value));
return indexContext.getIndexWriterConfig().getSimilarityFunction().compare(vector, queryVector);
}
Expand Down
Loading