benwtrent commented on code in PR #12160:
URL: https://github.com/apache/lucene/pull/12160#discussion_r1120747532


##########
lucene/core/src/java/org/apache/lucene/search/AbstractKnnVectorQuery.java:
##########
@@ -73,17 +77,48 @@ public Query rewrite(IndexSearcher indexSearcher) throws 
IOException {
               .build();
       Query rewritten = indexSearcher.rewrite(booleanQuery);
       filterWeight = indexSearcher.createWeight(rewritten, 
ScoreMode.COMPLETE_NO_SCORES, 1f);
+    } else {
+      filterWeight = null;
     }
 
-    for (LeafReaderContext ctx : reader.leaves()) {
-      TopDocs results = searchLeaf(ctx, filterWeight);
-      if (ctx.docBase > 0) {
-        for (ScoreDoc scoreDoc : results.scoreDocs) {
-          scoreDoc.doc += ctx.docBase;
-        }
-      }
-      perLeafResults[ctx.ord] = results;
-    }
+    List<FutureTask<TopDocs>> tasks =
+        reader.leaves().stream()
+            .map(
+                ctx ->
+                    new FutureTask<>(
+                        () -> {
+                          try {
+                            TopDocs results = searchLeaf(ctx, filterWeight);
+                            if (ctx.docBase > 0) {
+                              for (ScoreDoc scoreDoc : results.scoreDocs) {
+                                scoreDoc.doc += ctx.docBase;
+                              }
+                            }
+                            return results;
+                          } catch (IOException e) {
+                            throw new UncheckedIOException(e);
+                          }
+                        }))
+            .toList();
+
+    Executor executor = 
Objects.requireNonNullElse(indexSearcher.getExecutor(), Runnable::run);
+    SliceExecutor sliceExecutor = new SliceExecutor(executor);
+    sliceExecutor.invokeAll(tasks);

Review Comment:
   > That said, your suggestion aligns with how IndexSearcher currently works, 
so maybe we should apply it for now and discuss in a follow-up issue whether we 
should also delegate to the executor when there is a single segment.
   
   I am fine with that. 
   
   I think also that we could only check `indexSearcher.getExecutor() == null` 
instead of making a decision for the caller regarding the number of leaves.
   
   So, I would say for now only check if `indexSearcher.getExecutor() == null` 
and if it is, do it the old way.



-- 
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: issues-unsubscr...@lucene.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org

Reply via email to