AGNTLOG-2 - reduce payload memory usage in the logs pipeline#35351
Merged
dd-mergequeue[bot] merged 1 commit intomainfrom Mar 24, 2025
Merged
AGNTLOG-2 - reduce payload memory usage in the logs pipeline#35351dd-mergequeue[bot] merged 1 commit intomainfrom
dd-mergequeue[bot] merged 1 commit intomainfrom
Conversation
github-actions
bot
added
long review
ddrthall
added
the
qa/done
ddrthall
force-pushed
the
ryan.hall/reduce-log-payload-size
branch
from
fe1b36c to
1156c5f
Compare
Contributor
Uncompressed package size comparisonComparison with ancestor Diff per package
Decision |
Contributor
Test changes on VMUse this command from test-infra-definitions to manually test this PR changes on a VM: dda inv aws.create-vm --pipeline-id=59624210 --os-family=ubuntuNote: This applies to commit 530586f |
Contributor
Static quality checks ✅Please find below the results from static quality gates Successful checksInfo
|
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: 7e6f897 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | +0.87 | [+0.06, +1.68] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | +0.08 | [-0.69, +0.86] | 1 | Logs |
| ➖ | quality_gate_logs | % cpu utilization | +0.06 | [-2.71, +2.83] | 1 | Logs |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | +0.02 | [-0.66, +0.69] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http1 | egress throughput | +0.01 | [-0.79, +0.82] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.00 | [-0.03, +0.02] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | -0.01 | [-0.81, +0.79] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | -0.01 | [-0.64, +0.62] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | -0.01 | [-0.30, +0.28] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http2 | egress throughput | -0.02 | [-0.79, +0.76] | 1 | Logs |
| ➖ | file_tree | memory utilization | -0.05 | [-0.18, +0.09] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | -0.05 | [-0.51, +0.41] | 1 | Logs |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | -0.09 | [-0.87, +0.69] | 1 | Logs |
| ➖ | uds_dogstatsd_20mb_12k_contexts_20_senders | memory utilization | -0.28 | [-0.34, -0.22] | 1 | Logs |
| ➖ | quality_gate_idle | memory utilization | -0.87 | [-0.95, -0.78] | 1 | Logs bounds checks dashboard |
| ➖ | quality_gate_idle_all_features | memory utilization | -1.63 | [-1.75, -1.50] | 1 | Logs bounds checks dashboard |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | -2.62 | [-2.67, -2.56] | 1 | Logs |
Bounds Checks: ✅ Passed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ✅ | file_to_blackhole_0ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency_linear_load | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | memory_usage | 10/10 | |
| ✅ | quality_gate_idle | intake_connections | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | intake_connections | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_logs | intake_connections | 10/10 | |
| ✅ | quality_gate_logs | lost_bytes | 10/10 | |
| ✅ | quality_gate_logs | memory_usage | 10/10 |
Explanation
Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%
Performance changes are noted in the perf column of each table:
- ✅ = significantly better comparison variant performance
- ❌ = significantly worse comparison variant performance
- ➖ = no significant change in performance
A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".
For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:
-
Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
-
Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.
-
Its configuration does not mark it "erratic".
CI Pass/Fail Decision
✅ Passed. All Quality Gates passed.
- quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
- quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
- quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check lost_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
ddrthall
force-pushed
the
ryan.hall/reduce-log-payload-size
branch
3 times, most recently
from
2866296 to
f3b2642
Compare
ddrthall
force-pushed
the
ryan.hall/reduce-log-payload-size
branch
from
f3b2642 to
530586f
Compare
Contributor
Static quality checks ❌Please find below the results from static quality gates Error
Gate failure full details
Successful checksInfo
|
gh123man
reviewed
Mar 21, 2025
| func NewPayload(messages []*Message, encoded []byte, encoding string, unencodedSize int) *Payload { | ||
| messageMetas := make([]*MessageMetadata, len(messages)) | ||
| for i, m := range messages { | ||
| // Split the metadata from the message content to avoid holding the entire message in memory |
gh123man
reviewed
Mar 21, 2025
| message := NewMessage(content, nil, "", 2) | ||
|
|
||
| // Set up finalizer to track when content is GC'd | ||
| runtime.SetFinalizer(&message.content, trackGC) |
Member
There was a problem hiding this comment.
This is a cool test.
This seems like a nice way to assert memory behaviors in unit tests and possibly prevent memory regressions in the future 👍
hestonhoffman
approved these changes
Mar 21, 2025
gh123man
approved these changes
Mar 21, 2025
Contributor
Author
|
/merge |
|
View all feedbacks in Devflow UI.
The expected merge time in
|
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
3 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.
What does this PR do?
The logs payload entity stores metadata for all of the respective messages batched within, but also stores the message contents themselves. This isn't required, all consumers of the logs payload will interact with the encoded bytes buffer rather than the raw contents. Additionally, dropping these message contents on payload construction can save considerable amounts of RSS memory for log agents experiencing enough load to buffer multiple payloads.
This PR shallow copies the message metadata into each payload, rather than injecting the whole message entity.
Motivation
Describe how you validated your changes
Validation was completed by a combination of 1. unit test additions, 2. successful code compilation, 3. SMP performance runs
Possible Drawbacks / Trade-offs
Adding memory allocation in the critical path (for the metadata shallow copy) isn't ideal, but given we're already allocating a new array for the entirety of the contents in the same code path its impact shouldn't be noticeable.
Additional Notes
The memory savings for a system under heavy load is extreme. Attached is the memory comparison used in the file_to_blackhole_1000ms_latency_linear_load regression test:
