[#2275] Refactor Usage to address contention - #2276
Open
mattrpav wants to merge 2 commits into
Open
Conversation
JMH benchmark isolating the broker's topic send path: non-persistent
broker + NON_PERSISTENT BytesMessage, vm:// transport, async producer
send with unbounded window, producer flow control off, advisories/JMX/
scheduler/statistics/audit off, zero consumers. Parameters toggle a
shared topic vs a distinct topic per producer thread (distinctTopics)
and vm async dispatch vs direct dispatch on the calling thread
(vmAsync). Named *Benchmark so surefire skips it in CI.
Build (JDK 23+ needs proc=full for the JMH annotation processor):
mvn -pl activemq-unit-tests -am install -DskipTests -Dmaven.compiler.proc=full
Run (example: direct dispatch, both topic modes, thread sweep):
CP="activemq-unit-tests/target/test-classes:activemq-unit-tests/target/classes:$(mvn -q -pl activemq-unit-tests dependency:build-classpath -Dmdep.outputFile=/dev/stdout)"
java -cp "$CP" org.openjdk.jmh.Main TopicRawThroughputBenchmark \
-p numConsumers=0 -p useVirtualThread=false -p vmAsync=false \
-p distinctTopics=false,true -f 2 -wi 3 -w 2s -i 5 -r 2s \
-jvmArgs "-Xms4g -Xmx4g -XX:+AlwaysPreTouch"
On this commit (stock code) throughput DEGRADES as producer threads are
added, even with a topic per producer (e.g. ~5.2M msgs/sec at 1 thread
falling to ~1.1M at 22 threads on an 11-core machine). JMH stack
profiling attributes ~86% of lock wait time to the MemoryUsage
ReentrantReadWriteLock: every send takes exclusive write locks on a
usage hierarchy rooted in the broker-global SystemUsage, so destination
sharding cannot avoid the contention. The following commit removes it.
mattrpav
changed the title
mattrpav
marked this pull request as draft
August 2, 2026 11:59
mattrpav
changed the title
mattrpav
force-pushed
the
amq-gh-2275-usage-longadder
branch
from
August 2, 2026 21:57
ad3abff to
3ebd8d3
Compare
mattrpav
changed the title
mattrpav
marked this pull request as ready for review
August 3, 2026 01:50
mattrpav
changed the title
mattrpav
force-pushed
the
amq-gh-2275-usage-longadder
branch
from
August 3, 2026 18:22
3210d16 to
97286b9
Compare
cshannon
reviewed
Aug 3, 2026
cshannon
left a comment
cshannon
left a comment
Contributor
There was a problem hiding this comment.
Thanks for getting this started, I agree AtomicLong is probably the way to go for now.
It would be nice if we can get rid of the lock and calculate usage. Maybe we could compute the actual value (and not use percentage) on first use or startup (ie compute 70% of whatever the limit is) and just do long compare for checking if we hit the limit and then don't need to lock. I think we'd only need to lock if updating that computed value on change which would be rare (like JMX)
Replace the exclusive-lock counter in MemoryUsage with an AtomicLong and cache the current percentUsage bucket as absolute value bounds (PercentBounds) in base Usage. Mutations and reads compare the usage value against the cached bounds and only take the usageLock when a bound is crossed, where setPercentUsage recalculates the percent and bounds and fires the usual listener events and waitForSpace signals. The locked path recomputes after publishing until stable so a concurrent update cannot leave the published percent stale. Storage usage values change externally via store.size(), so base isFull(int) and getPercentUsage() use the same bounds check to skip the write lock on reads; the StoreUsage and TempUsage per-read recompute overrides are removed and JobSchedulerUsage now refreshes its percent on read.
mattrpav
force-pushed
the
amq-gh-2275-usage-longadder
branch
from
August 3, 2026 22:48
97286b9 to
47261e7
Compare
mattrpav
changed the title
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Reviewer notes:
I'm leaning towards AtomicLong as its more commonly understood and would make the change less brittle to maintain going forward.
edit: Design note: Consider calculating max limit and doing comparisons vs calculations