Add Lucene specific file extensions to core HybridFS

release-notes/opensearch-knn.release-notes-2.5.0.0.md

@@ -5,6 +5,7 @@ Compatible with OpenSearch 2.5.0
 ### Enhancements
 
 * Extend SystemIndexPlugin for k-NN model system index ([#630](https://github.com/opensearch-project/k-NN/pull/630))
+* Add Lucene specific file extensions to core HybridFS ([#721](https://github.com/opensearch-project/k-NN/pull/721))
 
 ### Bug Fixes
 

src/main/java/org/opensearch/knn/index/util/KNNEngine.java

@@ -12,6 +12,7 @@
 import org.opensearch.knn.index.KNNMethodContext;
 import org.opensearch.knn.index.SpaceType;
 
+import java.util.List;
 import java.util.Map;
 import java.util.Set;
 
@@ -176,4 +177,9 @@ public Boolean isInitialized() {
     public void setInitialized(Boolean isInitialized) {
         knnLibrary.setInitialized(isInitialized);
     }
+
+    @Override
+    public List<String> mmapFileExtensions() {
+        return knnLibrary.mmapFileExtensions();
+    }
 }

src/main/java/org/opensearch/knn/index/util/KNNLibrary.java

@@ -16,6 +16,8 @@
 import org.opensearch.knn.index.KNNMethodContext;
 import org.opensearch.knn.index.SpaceType;
 
+import java.util.Collections;
+import java.util.List;
 import java.util.Map;
 
 /**
@@ -113,4 +115,13 @@ public interface KNNLibrary {
      * @param isInitialized whether library has been initialized
      */
     void setInitialized(Boolean isInitialized);
+
+    /**
+     * Getter for mmap file extensions
+     *
+     * @return list of file extensions that will be read/write with mmap
+     */
+    default List<String> mmapFileExtensions() {
+        return Collections.EMPTY_LIST;
+    }
 }

src/main/java/org/opensearch/knn/index/util/Lucene.java

@@ -13,6 +13,7 @@
 import org.opensearch.knn.index.Parameter;
 import org.opensearch.knn.index.SpaceType;
 
+import java.util.List;
 import java.util.Map;
 
 import static org.opensearch.knn.common.KNNConstants.METHOD_HNSW;
@@ -73,4 +74,9 @@ public float score(float rawScore, SpaceType spaceType) {
         // score provided.
         return rawScore;
     }
+
+    @Override
+    public List<String> mmapFileExtensions() {
+        return List.of("vec", "vex");
+    }
 }

src/main/java/org/opensearch/knn/plugin/KNNPlugin.java

@@ -20,6 +20,7 @@
 import org.opensearch.knn.index.query.KNNWeight;
 import org.opensearch.knn.index.codec.KNNCodecService;
 import org.opensearch.knn.index.memory.NativeMemoryLoadStrategy;
+import org.opensearch.knn.index.util.KNNEngine;
 import org.opensearch.knn.indices.ModelGraveyard;
 import org.opensearch.knn.indices.ModelCache;
 import org.opensearch.knn.indices.ModelDao;
@@ -104,6 +105,8 @@
 import java.util.Map;
 import java.util.Optional;
 import java.util.function.Supplier;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
 
 import static java.util.Collections.singletonList;
 import static org.opensearch.knn.common.KNNConstants.KNN_THREAD_POOL_PREFIX;
@@ -338,4 +341,25 @@ public List<NamedXContentRegistry.Entry> getNamedXContent() {
     public Collection<SystemIndexDescriptor> getSystemIndexDescriptors(Settings settings) {
         return ImmutableList.of(new SystemIndexDescriptor(MODEL_INDEX_NAME, "Index for storing models used for k-NN indices"));
     }
+
+    /**
+     * Plugin can provide additional node settings, that includes new settings or overrides for existing one from core.
+     *
+     * @return settings that are set by plugin
+     */
+    @Override
+    public Settings additionalSettings() {
+        // We add engine specific extensions to the core list for HybridFS store type. We read existing values
+        // and append ours because in core setting will be replaced by override.
+        // Values are set as cluster defaults and are used at index creation time. Index specific overrides will take priority over values
+        // that are set here.
+        final List<String> engineSettings = Arrays.stream(KNNEngine.values())
+            .flatMap(engine -> engine.mmapFileExtensions().stream())
+            .collect(Collectors.toList());
+        final List<String> combinedSettings = Stream.concat(
+            IndexModule.INDEX_STORE_HYBRID_MMAP_EXTENSIONS.getDefault(Settings.EMPTY).stream(),
+            engineSettings.stream()
+        ).collect(Collectors.toList());
+        return Settings.builder().putList(IndexModule.INDEX_STORE_HYBRID_MMAP_EXTENSIONS.getKey(), combinedSettings).build();
+    }
 }

src/test/java/org/opensearch/knn/index/util/KNNEngineTests.java

@@ -8,6 +8,10 @@
 import org.opensearch.knn.KNNTestCase;
 import org.opensearch.knn.common.KNNConstants;
 
+import java.util.Arrays;
+import java.util.List;
+import java.util.stream.Collectors;
+
 public class KNNEngineTests extends KNNTestCase {
     /**
      * Check that version from engine and library match
@@ -47,4 +51,14 @@ public void testGetEngineFromPath() {
         String invalidPath = "test.invalid";
         expectThrows(IllegalArgumentException.class, () -> KNNEngine.getEngineNameFromPath(invalidPath));
     }
+
+    public void testMmapFileExtensions() {
+        final List<String> mmapExtensions = Arrays.stream(KNNEngine.values())
+            .flatMap(engine -> engine.mmapFileExtensions().stream())
+            .collect(Collectors.toList());
+        assertNotNull(mmapExtensions);
+        final List<String> expectedSettings = List.of("vex", "vec");
+        assertTrue(expectedSettings.containsAll(mmapExtensions));
+        assertTrue(mmapExtensions.containsAll(expectedSettings));
+    }
 }

src/test/java/org/opensearch/knn/index/util/LuceneTests.java

@@ -15,6 +15,7 @@
 
 import java.io.IOException;
 import java.util.Collections;
+import java.util.List;
 import java.util.Map;
 
 import static org.opensearch.knn.common.KNNConstants.METHOD_HNSW;
@@ -114,4 +115,12 @@ public void testVersion() {
         Lucene luceneInstance = Lucene.INSTANCE;
         assertEquals(Version.LATEST.toString(), luceneInstance.getVersion());
     }
+
+    public void testMmapFileExtensions() {
+        final List<String> luceneMmapExtensions = Lucene.INSTANCE.mmapFileExtensions();
+        assertNotNull(luceneMmapExtensions);
+        final List<String> expectedSettings = List.of("vex", "vec");
+        assertTrue(expectedSettings.containsAll(luceneMmapExtensions));
+        assertTrue(luceneMmapExtensions.containsAll(expectedSettings));
+    }
 }