Skip to content

GH-3879: Add cache to optimize header match performance. #3934

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.

Sign up for GitHub

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

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

GH-3879: Add cache to optimize header match performance. #3934

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fb754cc
spring-projectsGH-3879: Add cache to optimize header match performance.
chickenchickenlove May 30, 2025
e55bd5e
Remove useless comment.
chickenchickenlove May 30, 2025
390f4bf
Addressing PR review
chickenchickenlove May 30, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
43 changes: 43 additions & 0 deletions spring-kafka/src/main/java/org/springframework/kafka/support/AbstractKafkaHeaderMapper.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,7 @@
import org.springframework.core.log.LogAccessor;
import org.springframework.messaging.MessageHeaders;
import org.springframework.util.Assert;
import org.springframework.util.ConcurrentLruCache;
import org.springframework.util.ObjectUtils;
import org.springframework.util.PatternMatchUtils;

Expand Down Expand Up @@ -67,6 +68,8 @@ public abstract class AbstractKafkaHeaderMapper implements KafkaHeaderMapper {

private final List<HeaderMatcher> multiValueHeaderMatchers = new ArrayList<>();

private final HeaderPatternMatchCache headerMatchedCache = new HeaderPatternMatchCache();

private final Map<String, Boolean> rawMappedHeaders = new HashMap<>();

{
Expand Down Expand Up @@ -272,11 +275,22 @@ protected Object headerValueToAddOut(String key, Object value) {
* @since 4.0
*/
protected boolean doesMatchMultiValueHeader(String headerName) {
if (this.headerMatchedCache.isMultiValuePattern(headerName)) {
return true;
}

if (this.headerMatchedCache.isSingleValuePattern(headerName)) {
return false;
}

for (HeaderMatcher headerMatcher : this.multiValueHeaderMatchers) {
if (headerMatcher.matchHeader(headerName)) {
this.headerMatchedCache.cacheAsMultiValueHeader(headerName);
return true;
}
}

this.headerMatchedCache.cacheAsSingleValueHeader(headerName);
return false;
}

Expand Down Expand Up @@ -427,4 +441,33 @@ public boolean isNegated() {

}

/**
* A Cache that remembers whether a header name matches the multi-value pattern.
*/
private static final class HeaderPatternMatchCache {

private static final int MAX_SIZE = 1000;

private final ConcurrentLruCache<String, Boolean> multiValueHeaderPatternMatchCache = new ConcurrentLruCache<>(MAX_SIZE, key -> Boolean.TRUE);

private final ConcurrentLruCache<String, Boolean> singleValueHeaderPatternMatchCache = new ConcurrentLruCache<>(MAX_SIZE, key -> Boolean.TRUE);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the update .
Now I have a question: why no logic like “if not multi-value , than it is single”?
I just said something here about memory , and realized that we do waste it with this map for single-value headers. While it feels like just worry about caching multi is enough.
Looks like we need to dedicate more engineering to this algorithm to come up with better utilization.
Does it make sense what I’m asking?
Thanks.

Copy link
Contributor Author

@chickenchickenlove chickenchickenlove May 31, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The current implementation is designed to cache the result (either single-value or multi-value) for any header name that has undergone pattern matching at least once.

If we only store the result for multi-value headers, then even for header names that were already evaluated as single-value, the pattern matching logic would still be executed repeatedly each time.

Initially, I intended to use a single ConcurrentLruCache to store false for single-value and true for multi-value headers, but that behavior wasn’t supported🥲, which is why the current structure was chosen.

IMHO, for an effective pattern match cache, both outcomes—single-value and multi-value—should be stored. Currently, each ConcurrentLruCache is configured with a size of 1000. What do you think about reducing it to 500 to optimize memory usage?


boolean isMultiValuePattern(String headerName) {
return this.multiValueHeaderPatternMatchCache.contains(headerName);
}

boolean isSingleValuePattern(String headerName) {
return this.singleValueHeaderPatternMatchCache.contains(headerName);
}

void cacheAsSingleValueHeader(String headerName) {
this.singleValueHeaderPatternMatchCache.get(headerName);
}

void cacheAsMultiValueHeader(String headerName) {
this.multiValueHeaderPatternMatchCache.get(headerName);
}

}

}
98 changes: 98 additions & 0 deletions ...-kafka/src/test/java/org/springframework/kafka/support/DefaultKafkaHeaderMapperTests.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,8 @@
import org.apache.kafka.common.header.internals.RecordHeaders;
import org.assertj.core.api.InstanceOfAssertFactories;
import org.junit.jupiter.api.Test;
import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.ValueSource;

import org.springframework.core.log.LogAccessor;
import org.springframework.kafka.retrytopic.RetryTopicHeaders;
Expand Down Expand Up @@ -413,6 +415,102 @@ void multiValueHeaderToTest() {
.containsExactly(multiValueWildCardHeader2Value1, multiValueWildCardHeader2Value2);
}

@ParameterizedTest
@ValueSource(ints = {500, 1000, 2000})
void hugeNumberOfSingleValueHeaderToTest(int numberOfSingleValueHeaderCount) {
// GIVEN
Headers rawHeaders = new RecordHeaders();

String multiValueHeader1 = "test-multi-value1";
byte[] multiValueHeader1Value1 = { 0, 0, 0, 0 };
byte[] multiValueHeader1Value2 = { 0, 0, 0, 1 };

rawHeaders.add(multiValueHeader1, multiValueHeader1Value1);
rawHeaders.add(multiValueHeader1, multiValueHeader1Value2);

byte[] deliveryAttemptHeaderValue = { 0, 0, 0, 1 };
byte[] originalOffsetHeaderValue = { 0, 0, 0, 2 };
byte[] defaultHeaderAttemptsValues = { 0, 0, 0, 5 };

rawHeaders.add(KafkaHeaders.DELIVERY_ATTEMPT, deliveryAttemptHeaderValue);
rawHeaders.add(KafkaHeaders.ORIGINAL_OFFSET, originalOffsetHeaderValue);
rawHeaders.add(RetryTopicHeaders.DEFAULT_HEADER_ATTEMPTS, defaultHeaderAttemptsValues);

byte[] singleValueHeaderValue = { 0, 0, 0, 6 };
for (int i = 0; i < numberOfSingleValueHeaderCount; i++) {
String singleValueHeader = "test-single-value" + i;
rawHeaders.add(singleValueHeader, singleValueHeaderValue);
}

DefaultKafkaHeaderMapper mapper = new DefaultKafkaHeaderMapper();
mapper.setMultiValueHeaderPatterns(multiValueHeader1);

// WHEN
Map<String, Object> mappedHeaders = new HashMap<>();
mapper.toHeaders(rawHeaders, mappedHeaders);

// THEN
assertThat(mappedHeaders.get(KafkaHeaders.DELIVERY_ATTEMPT)).isEqualTo(1);
assertThat(mappedHeaders.get(KafkaHeaders.ORIGINAL_OFFSET)).isEqualTo(originalOffsetHeaderValue);
assertThat(mappedHeaders.get(RetryTopicHeaders.DEFAULT_HEADER_ATTEMPTS)).isEqualTo(defaultHeaderAttemptsValues);

for (int i = 0; i < numberOfSingleValueHeaderCount; i++) {
String singleValueHeader = "test-single-value" + i;
assertThat(mappedHeaders.get(singleValueHeader)).isEqualTo(singleValueHeaderValue);
}

assertThat(mappedHeaders)
.extractingByKey(multiValueHeader1, InstanceOfAssertFactories.list(byte[].class))
.containsExactly(multiValueHeader1Value1, multiValueHeader1Value2);
}

@ParameterizedTest
@ValueSource(ints = {500, 1000, 2000})
void hugeNumberOfMultiValueHeaderToTest(int numberOfMultiValueHeaderCount) {
// GIVEN
DefaultKafkaHeaderMapper mapper = new DefaultKafkaHeaderMapper();
Headers rawHeaders = new RecordHeaders();

byte[] multiValueHeader1Value1 = { 0, 0, 0, 0 };
byte[] multiValueHeader1Value2 = { 0, 0, 0, 1 };

for (int i = 0; i < numberOfMultiValueHeaderCount; i++) {
String multiValueHeader = "test-multi-value" + i;
mapper.setMultiValueHeaderPatterns(multiValueHeader);
rawHeaders.add(multiValueHeader, multiValueHeader1Value1);
rawHeaders.add(multiValueHeader, multiValueHeader1Value2);
}

byte[] deliveryAttemptHeaderValue = { 0, 0, 0, 1 };
byte[] originalOffsetHeaderValue = { 0, 0, 0, 2 };
byte[] defaultHeaderAttemptsValues = { 0, 0, 0, 5 };

rawHeaders.add(KafkaHeaders.DELIVERY_ATTEMPT, deliveryAttemptHeaderValue);
rawHeaders.add(KafkaHeaders.ORIGINAL_OFFSET, originalOffsetHeaderValue);
rawHeaders.add(RetryTopicHeaders.DEFAULT_HEADER_ATTEMPTS, defaultHeaderAttemptsValues);

String singleValueHeader = "test-single-value";
byte[] singleValueHeaderValue = { 0, 0, 0, 6 };
rawHeaders.add(singleValueHeader, singleValueHeaderValue);

// WHEN
Map<String, Object> mappedHeaders = new HashMap<>();
mapper.toHeaders(rawHeaders, mappedHeaders);

// THEN
assertThat(mappedHeaders.get(KafkaHeaders.DELIVERY_ATTEMPT)).isEqualTo(1);
assertThat(mappedHeaders.get(KafkaHeaders.ORIGINAL_OFFSET)).isEqualTo(originalOffsetHeaderValue);
assertThat(mappedHeaders.get(RetryTopicHeaders.DEFAULT_HEADER_ATTEMPTS)).isEqualTo(defaultHeaderAttemptsValues);
assertThat(mappedHeaders.get(singleValueHeader)).isEqualTo(singleValueHeaderValue);

for (int i = 0; i < numberOfMultiValueHeaderCount; i++) {
String multiValueHeader = "test-multi-value" + i;
assertThat(mappedHeaders)
.extractingByKey(multiValueHeader, InstanceOfAssertFactories.list(byte[].class))
.containsExactly(multiValueHeader1Value1, multiValueHeader1Value2);
}
}

@Test
void multiValueHeaderFromTest() {
// GIVEN
Expand Down