forked from pgjdbc/pgjdbc
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
perf: optimize Set<Integer> which are used for checking if oid should…
… be transferred with binary or text It looks like BitSet is faster than HashSet, so let's use it. Unfortunately, BitSet would consume memory in case the Oid value is high, and Oids can be up to 32bit, so we limit the max OID stored in a BitSet with 8192*8, so we effectively limit the extra per-connection memory by 16KiB (8KiB for send and 8KiB for receive)
- Loading branch information
Showing
5 changed files
with
385 additions
and
5 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
152 changes: 152 additions & 0 deletions
152
benchmarks/src/jmh/java/org/postgresql/benchmark/util/IntSetBenchmark.java
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,152 @@ | ||
| /* | ||
| * Copyright (c) 2024, PostgreSQL Global Development Group | ||
| * See the LICENSE file in the project root for more information. | ||
| */ | ||
|
|
||
| package org.postgresql.benchmark.util; | ||
|
|
||
| import org.postgresql.core.Oid; | ||
| import org.postgresql.util.internal.IntSet; | ||
|
|
||
| import it.unimi.dsi.fastutil.ints.IntOpenHashSet; | ||
| import org.openjdk.jmh.annotations.Benchmark; | ||
| import org.openjdk.jmh.annotations.BenchmarkMode; | ||
| import org.openjdk.jmh.annotations.Fork; | ||
| import org.openjdk.jmh.annotations.Measurement; | ||
| import org.openjdk.jmh.annotations.Mode; | ||
| import org.openjdk.jmh.annotations.OutputTimeUnit; | ||
| import org.openjdk.jmh.annotations.Param; | ||
| import org.openjdk.jmh.annotations.Scope; | ||
| import org.openjdk.jmh.annotations.Setup; | ||
| import org.openjdk.jmh.annotations.State; | ||
| import org.openjdk.jmh.annotations.Threads; | ||
| import org.openjdk.jmh.annotations.Warmup; | ||
| import org.openjdk.jmh.runner.Runner; | ||
| import org.openjdk.jmh.runner.RunnerException; | ||
| import org.openjdk.jmh.runner.options.Options; | ||
| import org.openjdk.jmh.runner.options.OptionsBuilder; | ||
| import org.roaringbitmap.RoaringBitmap; | ||
|
|
||
| import java.util.Arrays; | ||
| import java.util.BitSet; | ||
| import java.util.Collection; | ||
| import java.util.HashSet; | ||
| import java.util.Set; | ||
| import java.util.concurrent.TimeUnit; | ||
|
|
||
| /** | ||
| * Tests the performance of hex encoding. The results for OpenJDK 17.0.10 are as follows, so it | ||
| * looks like {@link Character#forDigit(int, int)} is the way to go. | ||
| * Here's a result for Apple M1 Max, Java 17.0.10, ARM64. | ||
| * 23 is present in the set, 73 and 123456 are both absent. | ||
| * <pre> | ||
| * Benchmark (oid) Mode Cnt Score Error Units | ||
| * IntSetBenchmark.bitSet_contains 23 avgt 15 0,984 ± 0,009 ns/op | ||
| * IntSetBenchmark.bitSet_contains 73 avgt 15 0,979 ± 0,005 ns/op | ||
| * IntSetBenchmark.bitSet_contains 123456 avgt 15 0,688 ± 0,015 ns/op | ||
| * IntSetBenchmark.hashSet_contains 23 avgt 15 2,026 ± 0,013 ns/op | ||
| * IntSetBenchmark.hashSet_contains 73 avgt 15 1,985 ± 0,004 ns/op | ||
| * IntSetBenchmark.hashSet_contains 123456 avgt 15 1,968 ± 0,032 ns/op | ||
| * IntSetBenchmark.intOpenSet_contains 23 avgt 15 1,015 ± 0,011 ns/op | ||
| * IntSetBenchmark.intOpenSet_contains 73 avgt 15 5,720 ± 0,596 ns/op | ||
| * IntSetBenchmark.intOpenSet_contains 123456 avgt 15 8,430 ± 0,007 ns/op | ||
| * IntSetBenchmark.intSet_contains 23 avgt 15 1,101 ± 0,009 ns/op | ||
| * IntSetBenchmark.intSet_contains 73 avgt 15 1,117 ± 0,005 ns/op | ||
| * IntSetBenchmark.intSet_contains 123456 avgt 15 0,693 ± 0,010 ns/op | ||
| * IntSetBenchmark.roaring_contains 23 avgt 15 5,012 ± 0,044 ns/op | ||
| * IntSetBenchmark.roaring_contains 73 avgt 15 5,561 ± 0,077 ns/op | ||
| * IntSetBenchmark.roaring_contains 123456 avgt 15 1,163 ± 0,012 ns/op | ||
| * </pre> | ||
| */ | ||
| @Fork(value = 3, jvmArgsPrepend = "-Xmx128m") | ||
| @Measurement(iterations = 5, time = 1, timeUnit = TimeUnit.SECONDS) | ||
| @Warmup(iterations = 5, time = 1, timeUnit = TimeUnit.SECONDS) | ||
| @State(Scope.Thread) | ||
| @Threads(1) | ||
| @BenchmarkMode(Mode.AverageTime) | ||
| @OutputTimeUnit(TimeUnit.NANOSECONDS) | ||
| public class IntSetBenchmark { | ||
| /** | ||
| * See {@code PgConnection#getSupportedBinaryOids()}. | ||
| */ | ||
| private static final Collection<? extends Integer> SUPPORTED_BINARY_OIDS = Arrays.asList( | ||
| Oid.BYTEA, | ||
| Oid.INT2, | ||
| Oid.INT4, | ||
| Oid.INT8, | ||
| Oid.FLOAT4, | ||
| Oid.FLOAT8, | ||
| Oid.NUMERIC, | ||
| Oid.TIME, | ||
| Oid.DATE, | ||
| Oid.TIMETZ, | ||
| Oid.TIMESTAMP, | ||
| Oid.TIMESTAMPTZ, | ||
| Oid.BYTEA_ARRAY, | ||
| Oid.INT2_ARRAY, | ||
| Oid.INT4_ARRAY, | ||
| Oid.INT8_ARRAY, | ||
| Oid.OID_ARRAY, | ||
| Oid.FLOAT4_ARRAY, | ||
| Oid.FLOAT8_ARRAY, | ||
| Oid.VARCHAR_ARRAY, | ||
| Oid.TEXT_ARRAY, | ||
| Oid.POINT, | ||
| Oid.BOX, | ||
| Oid.UUID); | ||
|
|
||
| IntSet intSet = new IntSet(); | ||
| BitSet bitSet = new BitSet(); | ||
| Set<Integer> hashSet = new HashSet<>(); | ||
| RoaringBitmap roaringBitmap = new RoaringBitmap(); | ||
| IntOpenHashSet intOpenHashSet = new IntOpenHashSet(); | ||
|
|
||
| @Param({"23", "73", "123456"}) | ||
| int oid; | ||
|
|
||
| @Setup | ||
| public void setup() { | ||
| intSet.addAll(SUPPORTED_BINARY_OIDS); | ||
| hashSet.addAll(SUPPORTED_BINARY_OIDS); | ||
| for (Integer oid : SUPPORTED_BINARY_OIDS) { | ||
| bitSet.set(oid); | ||
| roaringBitmap.add(oid); | ||
| intOpenHashSet.add((int) oid); | ||
| } | ||
| } | ||
|
|
||
| @Benchmark | ||
| public boolean intSet_contains() { | ||
| return intSet.contains(oid); | ||
| } | ||
|
|
||
| @Benchmark | ||
| public boolean bitSet_contains() { | ||
| return bitSet.get(oid); | ||
| } | ||
|
|
||
| @Benchmark | ||
| public boolean hashSet_contains() { | ||
| return hashSet.contains(oid); | ||
| } | ||
|
|
||
| @Benchmark | ||
| public boolean roaring_contains() { | ||
| return roaringBitmap.contains(oid); | ||
| } | ||
|
|
||
| @Benchmark | ||
| public boolean intOpenSet_contains() { | ||
| return intOpenHashSet.contains(oid); | ||
| } | ||
|
|
||
| public static void main(String[] args) throws RunnerException { | ||
| Options opt = new OptionsBuilder() | ||
| .include(IntSetBenchmark.class.getSimpleName()) | ||
| //.addProfiler(GCProfiler.class) | ||
| .detectJvmArgs() | ||
| .build(); | ||
|
|
||
| new Runner(opt).run(); | ||
| } | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
126 changes: 126 additions & 0 deletions
126
pgjdbc/src/main/java/org/postgresql/util/internal/IntSet.java
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,126 @@ | ||
| /* | ||
| * Copyright (c) 2024, PostgreSQL Global Development Group | ||
| * See the LICENSE file in the project root for more information. | ||
| */ | ||
|
|
||
| package org.postgresql.util.internal; | ||
|
|
||
| import org.postgresql.core.Oid; | ||
|
|
||
| import org.checkerframework.checker.nullness.qual.Nullable; | ||
|
|
||
| import java.util.BitSet; | ||
| import java.util.Collection; | ||
| import java.util.HashSet; | ||
| import java.util.Set; | ||
|
|
||
| /** | ||
| * Read-optimized {@code Set} for storing {@link Oid} values. | ||
| * Note: the set does not support nullable values. | ||
| */ | ||
| public final class IntSet { | ||
| /** | ||
| * Maximal Oid that will bs stored in {@link BitSet}. | ||
| * If Oid exceeds this value, then it will be stored in {@code Set<Int>} only. | ||
| * In theory, Oids can be up to 32bit, so we want to limit per-connection memory utilization. | ||
| * Allow {@code BitSet} to consume up to 8KiB (one for send and one for receive). | ||
| */ | ||
| private static final int MAX_OID_TO_STORE_IN_BITSET = 8192 * 8; | ||
|
|
||
| /** | ||
| * Contains values outside [0..MAX_OID_TO_STORE_IN_BITSET] range. | ||
| * This field is null if bitSet contains all the values. | ||
| */ | ||
| private @Nullable Set<Integer> set; | ||
|
|
||
| /** | ||
| * Contains values in range of [0..MAX_OID_TO_STORE_IN_BITSET]. | ||
| */ | ||
| private final BitSet bitSet = new BitSet(); | ||
|
|
||
| /** | ||
| * Clears the contents of the set. | ||
| */ | ||
| public void clear() { | ||
| set = null; | ||
| bitSet.clear(); | ||
| } | ||
|
|
||
| /** | ||
| * Adds all the values to the set. | ||
| * @param values set of values to add | ||
| */ | ||
| public void addAll(Collection<? extends Integer> values) { | ||
| for (Integer value : values) { | ||
| add(value); | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Adds a single value to the set. | ||
| * | ||
| * @param value value to add | ||
| * @return true if the set did not already contain the specified value | ||
| */ | ||
| public boolean add(int value) { | ||
| if (value >= 0 && value <= MAX_OID_TO_STORE_IN_BITSET) { | ||
| boolean contains = bitSet.get(value); | ||
| if (!contains) { | ||
| bitSet.set(value); | ||
| return true; | ||
| } | ||
| return false; | ||
| } | ||
| Set<Integer> set = this.set; | ||
| if (set == null) { | ||
| this.set = set = new HashSet<>(); | ||
| } | ||
| return set.add(value); | ||
| } | ||
|
|
||
| /** | ||
| * Removes a value from the set. | ||
| * @param value value to remove | ||
| * @return true if the element was | ||
| */ | ||
| public boolean remove(int value) { | ||
| if (value >= 0 && value <= MAX_OID_TO_STORE_IN_BITSET) { | ||
| boolean contains = bitSet.get(value); | ||
| if (contains) { | ||
| bitSet.clear(value); | ||
| return true; | ||
| } | ||
| return false; | ||
| } | ||
| Set<Integer> set = this.set; | ||
| return set != null && set.remove(value); | ||
| } | ||
|
|
||
| /** | ||
| * Checks if a given value belongs to the set. | ||
| * @param value value to check | ||
| * @return true if the value belons to the set | ||
| */ | ||
| public boolean contains(int value) { | ||
| if (value >= 0 && value <= MAX_OID_TO_STORE_IN_BITSET) { | ||
| return bitSet.get(value); | ||
| } | ||
| Set<Integer> set = this.set; | ||
| return set != null && set.contains(value); | ||
| } | ||
|
|
||
| /** | ||
| * Returns a mutable snapshot of the values stored in the current set. | ||
| * @return a mutable snapshot of the values stored in the current set | ||
| */ | ||
| public Set<Integer> toMutableSet() { | ||
| Set<Integer> set = this.set; | ||
| Set<Integer> result = new HashSet<>( | ||
| (int) ((bitSet.cardinality() + (set != null ? set.size() : 0)) / 0.75f)); | ||
| if (set != null) { | ||
| result.addAll(set); | ||
| } | ||
| bitSet.stream().forEach(result::add); | ||
| return result; | ||
| } | ||
| } |
Oops, something went wrong.