experiment(datadog metrics): build requests on global thread pool #413
Conversation
github-actions
bot
added
area/components
This stack of pull requests is managed by Graphite. Learn more about stacking. |
Regression Detector (DogStatsD)Regression Detector ResultsRun ID: adcd4def-37f4-487e-a96b-a42d525d06b0 Baseline: 7.65.0-rc.9 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | dsd_uds_100mb_3k_contexts_distributions_only | memory utilization | +0.50 | [+0.29, +0.70] | 1 | |
| ➖ | dsd_uds_100mb_3k_contexts | ingress throughput | +0.01 | [-0.09, +0.11] | 1 | |
| ➖ | dsd_uds_100mb_250k_contexts | ingress throughput | +0.00 | [-0.10, +0.10] | 1 | |
| ➖ | dsd_uds_40mb_12k_contexts_40_senders | ingress throughput | +0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_1mb_50k_contexts | ingress throughput | +0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_1mb_50k_contexts_memlimit | ingress throughput | -0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_512kb_3k_contexts | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | |
| ➖ | dsd_uds_1mb_3k_contexts_dualship | ingress throughput | -0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_1mb_3k_contexts | ingress throughput | -0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_10mb_3k_contexts | ingress throughput | -0.00 | [-0.03, +0.03] | 1 | |
| ➖ | quality_gates_idle_rss | memory utilization | -1.60 | [-1.74, -1.46] | 1 | |
| ➖ | dsd_uds_500mb_3k_contexts | ingress throughput | -3.05 | [-3.12, -2.97] | 1 |
Bounds Checks: ❌ Failed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ❌ | quality_gates_idle_rss | memory_usage | 0/10 |
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".
Regression Detector (Saluki)Regression Detector ResultsRun ID: c04b27c5-c65c-4b3b-9569-8e860186c9b8 Baseline: bd6799f Optimization Goals: ✅ Improvement(s) detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ✅ | dsd_uds_1mb_50k_contexts_memlimit | ingress throughput | +17.62 | [+15.95, +19.28] | 1 | |
| ➖ | dsd_uds_500mb_3k_contexts | ingress throughput | +1.80 | [+1.68, +1.92] | 1 | |
| ➖ | dsd_uds_100mb_250k_contexts | ingress throughput | +0.01 | [-0.07, +0.09] | 1 | |
| ➖ | dsd_uds_50mb_10k_contexts_no_inlining | ingress throughput | +0.01 | [-0.10, +0.11] | 1 | |
| ➖ | dsd_uds_1mb_50k_contexts | ingress throughput | +0.00 | [-0.00, +0.01] | 1 | |
| ➖ | dsd_uds_50mb_10k_contexts_no_inlining_no_allocs | ingress throughput | +0.00 | [-0.09, +0.10] | 1 | |
| ➖ | dsd_uds_40mb_12k_contexts_40_senders | ingress throughput | +0.00 | [-0.03, +0.03] | 1 | |
| ➖ | dsd_uds_512kb_3k_contexts | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | |
| ➖ | dsd_uds_100mb_3k_contexts | ingress throughput | -0.00 | [-0.06, +0.06] | 1 | |
| ➖ | dsd_uds_1mb_3k_contexts | ingress throughput | -0.00 | [-0.02, +0.01] | 1 | |
| ➖ | dsd_uds_1mb_3k_contexts_dualship | ingress throughput | -0.01 | [-0.02, +0.01] | 1 | |
| ➖ | dsd_uds_10mb_3k_contexts | ingress throughput | -0.02 | [-0.05, +0.01] | 1 | |
| ➖ | dsd_uds_100mb_3k_contexts_distributions_only | memory utilization | -3.67 | [-3.83, -3.51] | 1 | |
| ✅ | quality_gates_idle_rss | memory utilization | -5.51 | [-5.65, -5.37] | 1 |
Bounds Checks: ✅ Passed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ✅ | quality_gates_idle_rss | 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".
Regression Detector LinksExperiment Result Links
|
tobz
force-pushed
the
tobz/generic-async-thread-pool
branch
2 times, most recently
from
bfd9cb4 to
9d1396e
Compare
tobz
force-pushed
the
tobz/dd-metrics-request-builder-thread-poolified
branch
from
fb1c72e to
c67420e
Compare
tobz
changed the title
tobz
changed the base branch from
tobz/generic-async-thread-pool
to
graphite-base/413
tobz
force-pushed
the
tobz/dd-metrics-request-builder-thread-poolified
branch
from
4c0a340 to
0909e87
Compare
Summary
This PR updates the Datadog Metrics destination to move request building to the global thread pool, in an attempt to reduce the amount of synchronous code (Protocol Buffers encoding, compression) happening on the primary runtime.
Our hypothesis is that by doing so, we can greatly reduce the tail latency of other tasks on the primary runtime, such as socket handler tasks, and in doing so, improve the client-perceived latency of interacting with ADP.
Change Type
How did you test this PR?
N/A
References
N/A