Решетников Алексей, Магистратура ИТМО "Распределенные веб-сервисы", sampled index

src/main/java/ru/vk/itmo/reference/SSTable.java

@@ -5,10 +5,14 @@
 
 import java.lang.foreign.MemorySegment;
 import java.lang.foreign.ValueLayout;
+import java.util.ArrayList;
 import java.util.Collections;
 import java.util.Iterator;
+import java.util.List;
 import java.util.NoSuchElementException;
 
+import static ru.vk.itmo.reference.SampledIndex.SAMPLED_INDEX_STEP;
+
 /**
  * Persistent SSTable in data file and index file.
  *
@@ -19,6 +23,7 @@ final class SSTable {
     final int sequence;
 
     private final MemorySegment index;
+    private final List<SampledIndex> sampledIndices;
     private final MemorySegment data;
     private final long size;
 
@@ -30,6 +35,23 @@ final class SSTable {
         this.index = index;
         this.data = data;
         this.size = index.byteSize() / Long.BYTES;
+
+        this.sampledIndices = new ArrayList<>();
+        for (int indexOffset = 0; indexOffset < size; indexOffset += SAMPLED_INDEX_STEP) {
+            // | [kv][kv][kv][kv][kv][kv][kv] | [kv][kv][kv][kv]...[kv]
+            //    ^
+            //    |
+            // keyAtTheStart
+            final long keyAtTheStartOffset = entryOffset(indexOffset);
+            final long keyAtTheStartLength = getLength(keyAtTheStartOffset);
+
+            sampledIndices.add(
+                    new SampledIndex(
+                            keyAtTheStartOffset + Long.BYTES, keyAtTheStartLength,
+                            indexOffset
+                    )
+            );
+        }
     }
 
     SSTable withSequence(final int sequence) {
@@ -39,6 +61,33 @@ SSTable withSequence(final int sequence) {
                 data);
     }
 
+    private long sampledIndexBinarySearch(final MemorySegment key) {
+        int low = 0;
+        int high = sampledIndices.size() - 1;
+
+        while (low <= high) {
+            final int mid = (low + high) >>> 1;
+            final SampledIndex sampledIndex = sampledIndices.get(mid);
+
+            final long keyAtTheStartOffset = sampledIndex.keyAtTheStartOffset();
+            final long keyAtTheStartLength = sampledIndex.keyAtTheStartLength();
+            final int compare = MemorySegmentComparator.compare(
+                    data, keyAtTheStartOffset, keyAtTheStartLength,
+                    key, 0L, key.byteSize()
+            );
+
+            if (compare < 0) {
+                low = mid + 1;
+            } else if (compare > 0) {
+                high = mid - 1;
+            } else {
+                return sampledIndex.offset();
+            }
+        }
+
+        return sampledIndices.get(high).offset();
+    }
+
     /**
      * Returns index of the entry if found; otherwise, (-(insertion point) - 1).
      * The insertion point is defined as the point at which the key would be inserted:
@@ -48,8 +97,8 @@ SSTable withSequence(final int sequence) {
      * if and only if the key is found.
      */
     private long entryBinarySearch(final MemorySegment key) {
-        long low = 0L;
-        long high = size - 1;
+        long low = sampledIndexBinarySearch(key);
+        long high = low + SAMPLED_INDEX_STEP;
 
         while (low <= high) {
             final long mid = (low + high) >>> 1;

src/main/java/ru/vk/itmo/reference/SampledIndex.java

@@ -0,0 +1,9 @@
+package ru.vk.itmo.reference;
+
+public record SampledIndex(
+        long keyAtTheStartOffset,
+        long keyAtTheStartLength,
+        long offset
+) {
+    static final int SAMPLED_INDEX_STEP = 16;
+}

src/main/java/ru/vk/itmo/test/reference/ReferenceDaoFactory.java

@@ -16,7 +16,7 @@
  *
  * @author incubos
  */
-@DaoFactory(stage = 5)
+@DaoFactory(stage = 500)
 public class ReferenceDaoFactory implements DaoFactory.Factory<MemorySegment, Entry<MemorySegment>> {
     @Override
     public Dao<MemorySegment, Entry<MemorySegment>> createDao(final Config config) throws IOException {