-
Notifications
You must be signed in to change notification settings - Fork 24.9k
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.
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
Speed up writeVInt #62345
Speed up writeVInt #62345
Changes from 5 commits
15471e6
c65a0b1
5cd3531
929297f
ca6cd67
5beaac8
ec40216
e35e1c5
4f6d4d8
4d1d7bd
bd9cc57
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,92 @@ | ||
/* | ||
* Licensed to Elasticsearch under one or more contributor | ||
* license agreements. See the NOTICE file distributed with | ||
* this work for additional information regarding copyright | ||
* ownership. Elasticsearch licenses this file to you under | ||
* the Apache License, Version 2.0 (the "License"); you may | ||
* not use this file except in compliance with the License. | ||
* You may obtain a copy of the License at | ||
* | ||
* http://www.apache.org/licenses/LICENSE-2.0 | ||
* | ||
* Unless required by applicable law or agreed to in writing, | ||
* software distributed under the License is distributed on an | ||
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY | ||
* KIND, either express or implied. See the License for the | ||
* specific language governing permissions and limitations | ||
* under the License. | ||
*/ | ||
|
||
package org.elasticsearch.benchmark.search.aggregations.bucket.terms; | ||
|
||
import org.apache.lucene.util.BytesRef; | ||
import org.elasticsearch.common.io.stream.DelayableWriteable; | ||
import org.elasticsearch.common.io.stream.NamedWriteableRegistry; | ||
import org.elasticsearch.search.DocValueFormat; | ||
import org.elasticsearch.search.aggregations.BucketOrder; | ||
import org.elasticsearch.search.aggregations.InternalAggregation; | ||
import org.elasticsearch.search.aggregations.InternalAggregations; | ||
import org.elasticsearch.search.aggregations.bucket.terms.StringTerms; | ||
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.Warmup; | ||
|
||
import java.util.ArrayList; | ||
import java.util.List; | ||
import java.util.concurrent.TimeUnit; | ||
|
||
@Fork(2) | ||
@Warmup(iterations = 10) | ||
@Measurement(iterations = 5) | ||
@BenchmarkMode(Mode.AverageTime) | ||
@OutputTimeUnit(TimeUnit.MILLISECONDS) | ||
@State(Scope.Benchmark) | ||
public class StringTermsSerializationBenchmark { | ||
private static final NamedWriteableRegistry REGISTRY = new NamedWriteableRegistry( | ||
List.of(new NamedWriteableRegistry.Entry(InternalAggregation.class, StringTerms.NAME, StringTerms::new)) | ||
); | ||
@Param(value = { "1000" }) | ||
private int buckets; | ||
|
||
private DelayableWriteable<InternalAggregations> results; | ||
|
||
@Setup | ||
public void initResults() { | ||
results = DelayableWriteable.referencing(InternalAggregations.from(List.of(newTerms(true)))); | ||
} | ||
|
||
private StringTerms newTerms(boolean withNested) { | ||
List<StringTerms.Bucket> resultBuckets = new ArrayList<>(buckets); | ||
for (int i = 0; i < buckets; i++) { | ||
InternalAggregations inner = withNested ? InternalAggregations.from(List.of(newTerms(false))) : InternalAggregations.EMPTY; | ||
resultBuckets.add(new StringTerms.Bucket(new BytesRef("test" + i), i, inner, false, 0, DocValueFormat.RAW)); | ||
} | ||
return new StringTerms( | ||
"test", | ||
BucketOrder.key(true), | ||
BucketOrder.key(true), | ||
buckets, | ||
1, | ||
null, | ||
DocValueFormat.RAW, | ||
buckets, | ||
false, | ||
100000, | ||
resultBuckets, | ||
0 | ||
); | ||
} | ||
|
||
@Benchmark | ||
public DelayableWriteable<InternalAggregations> serialize() { | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I'm unsure if we actually want this benchmark, especially compared to the one that @jimczi showed me. But it is fairly targeted which can be useful. |
||
return results.asSerialized(InternalAggregations::readFrom, REGISTRY); | ||
} | ||
} |
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -218,14 +218,85 @@ public void writeInt(int i) throws IOException { | |
* using {@link #writeInt} | ||
*/ | ||
public void writeVInt(int i) throws IOException { | ||
final byte[] buffer = scratch.get(); | ||
int index = 0; | ||
while ((i & ~0x7F) != 0) { | ||
buffer[index++] = ((byte) ((i & 0x7f) | 0x80)); | ||
i >>>= 7; | ||
/* | ||
* Pick the number of bytes that we need based on the value and then | ||
* encode the int, unrolling the loops by hand. This allows writing | ||
* small numbers to use `writeByte` which is simple and fast. The | ||
* unrolling saves a few comparisons and bitwise operations. All | ||
* together this saves quite a bit of time compared to a naive | ||
* implementation. | ||
*/ | ||
switch (Integer.numberOfLeadingZeros(i)) { | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. This gets compiled to |
||
case 32: | ||
case 31: | ||
case 30: | ||
case 29: | ||
case 28: | ||
case 27: | ||
case 26: | ||
case 25: | ||
writeByte((byte) i); | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I love this until here :) The fact that we can special case Hard coding all possible offsets below and doing all the buffer getting and I benchmarked both this version and: public void writeVInt(int i) throws IOException {
if (Integer.numberOfLeadingZeros(i) > 24) {
writeByte((byte) i);
} else {
final byte[] buffer = scratch.get();
int index = 0;
do {
buffer[index++] = ((byte) ((i & 0x7f) | 0x80));
i >>>= 7;
} while ((i & ~0x7F) != 0);
buffer[index++] = ((byte) i);
writeBytes(buffer, 0, index);
}
} and I can't see statistically significant difference so that's not worth the complication IMO. I would in fact expect the above version with the loop to be faster than what is in this PR in the real world because the smaller method size has a better better chance of getting inlined in some places ( I suppose you could work around the code bloat by doing this: final int leadingZeros = Integer.numberOfLeadingZeros(i);
if (Integer.numberOfLeadingZeros(i) > 24) {
writeByte((byte) i);
} else {
final byte[] buffer = scratch.get();
final int length;
switch (leadingZeros) {
case 24:
case 23:
case 22:
case 21:
case 20:
case 19:
case 18:
buffer[0] = (byte) (i & 0x7f | 0x80);
buffer[1] = (byte) (i >>> 7);
assert buffer[1] <= 0x7f;
length = 2;
break;
case 17:
case 16:
case 15:
case 14:
case 13:
case 12:
case 11:
buffer[0] = (byte) (i & 0x7f | 0x80);
buffer[1] = (byte) ((i >>> 7) & 0x7f | 0x80);
buffer[2] = (byte) (i >>> 14);
assert buffer[2] <= 0x7f;
length = 3;
break;
case 10:
case 9:
case 8:
case 7:
case 6:
case 5:
case 4:
buffer[0] = (byte) (i & 0x7f | 0x80);
buffer[1] = (byte) ((i >>> 7) & 0x7f | 0x80);
buffer[2] = (byte) ((i >>> 14) & 0x7f | 0x80);
buffer[3] = (byte) (i >>> 21);
assert buffer[3] <= 0x7f;
length = 4;
break;
case 3:
case 2:
case 1:
case 0:
buffer[0] = (byte) (i & 0x7f | 0x80);
buffer[1] = (byte) ((i >>> 7) & 0x7f | 0x80);
buffer[2] = (byte) ((i >>> 14) & 0x7f | 0x80);
buffer[3] = (byte) ((i >>> 21) & 0x7f | 0x80);
buffer[4] = (byte) (i >>> 28);
assert buffer[4] <= 0x7f;
length = 5;
break;
default:
throw new UnsupportedOperationException(
"Can't encode [" + i + "]. Missing case for [" + Integer.numberOfLeadingZeros(i) + "]?"
);
}
writeBytes(buffer, 0, length);
} but I can't measure a performance difference to the loop at all so personally I'd go for the shorter loop just for simplicity's sake. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I think your right about your implementation being faster in practice. I put together a quick and dirty benchmark for writeVInt directly and my hand unrolled thing is faster there. By a pretty wide margin. But the benchmark for serializing the agg result is slower. I can see in the decompiled output that my method results in writeVInt not being inlined because it is too big like you say. And yours gets it inlined.
The compromise solution doesn't seem to shrink the method enough. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Thanks for testing this! |
||
return; | ||
case 24: | ||
case 23: | ||
case 22: | ||
case 21: | ||
case 20: | ||
case 19: | ||
case 18: | ||
byte[] buffer = scratch.get(); | ||
buffer[0] = (byte) (i & 0x7f | 0x80); | ||
buffer[1] = (byte) (i >>> 7); | ||
assert buffer[1] <= 0x7f; | ||
writeBytes(buffer, 0, 2); | ||
return; | ||
case 17: | ||
case 16: | ||
case 15: | ||
case 14: | ||
case 13: | ||
case 12: | ||
case 11: | ||
buffer = scratch.get(); | ||
buffer[0] = (byte) (i & 0x7f | 0x80); | ||
buffer[1] = (byte) ((i >>> 7) & 0x7f | 0x80); | ||
buffer[2] = (byte) (i >>> 14); | ||
assert buffer[2] <= 0x7f; | ||
writeBytes(buffer, 0, 3); | ||
return; | ||
case 10: | ||
case 9: | ||
case 8: | ||
case 7: | ||
case 6: | ||
case 5: | ||
case 4: | ||
buffer = scratch.get(); | ||
buffer[0] = (byte) (i & 0x7f | 0x80); | ||
buffer[1] = (byte) ((i >>> 7) & 0x7f | 0x80); | ||
buffer[2] = (byte) ((i >>> 14) & 0x7f | 0x80); | ||
buffer[3] = (byte) (i >>> 21); | ||
assert buffer[3] <= 0x7f; | ||
writeBytes(buffer, 0, 4); | ||
return; | ||
case 3: | ||
case 2: | ||
case 1: | ||
case 0: | ||
buffer = scratch.get(); | ||
buffer[0] = (byte) (i & 0x7f | 0x80); | ||
buffer[1] = (byte) ((i >>> 7) & 0x7f | 0x80); | ||
buffer[2] = (byte) ((i >>> 14) & 0x7f | 0x80); | ||
buffer[3] = (byte) ((i >>> 21) & 0x7f | 0x80); | ||
buffer[4] = (byte) (i >>> 28); | ||
assert buffer[4] <= 0x7f; | ||
writeBytes(buffer, 0, 5); | ||
return; | ||
default: | ||
throw new UnsupportedOperationException( | ||
"Can't encode [" + i + "]. Missing case for [" + Integer.numberOfLeadingZeros(i) + "]?" | ||
); | ||
} | ||
buffer[index++] = ((byte) i); | ||
writeBytes(buffer, 0, index); | ||
} | ||
|
||
/** | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This was just wrong.