enhancement: incorporate origin key into context key hash to shrink size of ContextKey
#428
Conversation
…ize of ContextKey
tobz
force-pushed
the
tobz/shrink-context-key-size
branch
from
f9b0aed to
ee396d5
Compare
Regression Detector (DogStatsD)Regression Detector ResultsRun ID: f22a284b-fe5a-4b96-97cb-0e3735117157 Baseline: 7.62.0-rc.2 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | quality_gates_idle_rss | memory utilization | +1.02 | [+0.94, +1.09] | 1 | |
| ➖ | dsd_uds_100mb_3k_contexts_distributions_only | memory utilization | +0.86 | [+0.65, +1.06] | 1 | |
| ➖ | dsd_uds_100mb_3k_contexts | ingress throughput | +0.02 | [-0.03, +0.06] | 1 | |
| ➖ | dsd_uds_1mb_50k_contexts_memlimit | ingress throughput | +0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_500mb_3k_contexts | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | |
| ➖ | dsd_uds_512kb_3k_contexts | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | |
| ➖ | dsd_uds_1mb_50k_contexts | ingress throughput | -0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_100mb_250k_contexts | ingress throughput | -0.00 | [-0.00, +0.00] | 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 | [-0.02, +0.01] | 1 | |
| ➖ | dsd_uds_40mb_12k_contexts_40_senders | ingress throughput | -0.00 | [-0.01, +0.00] | 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: 2c6f7660-bb7e-4e66-bca7-8ce0aaf59bce Baseline: 799f140 ❌ Experiments with missing or malformed dataThis is a critical error. No usable optimization goal data was produced by the listed experiments. This may be a result of misconfiguration. Ping #single-machine-performance and we can help out.
Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | dsd_uds_1mb_50k_contexts_memlimit | ingress throughput | +2.84 | [+2.36, +3.33] | 1 | |
| ➖ | dsd_uds_500mb_3k_contexts | ingress throughput | +1.66 | [+1.55, +1.76] | 1 | |
| ➖ | dsd_uds_100mb_3k_contexts_distributions_only | memory utilization | +0.23 | [+0.11, +0.36] | 1 | |
| ➖ | dsd_uds_50mb_10k_contexts_no_inlining | ingress throughput | +0.01 | [-0.05, +0.08] | 1 | |
| ➖ | dsd_uds_50mb_10k_contexts_no_inlining_no_allocs | ingress throughput | +0.01 | [-0.04, +0.05] | 1 | |
| ➖ | dsd_uds_40mb_12k_contexts_40_senders | ingress throughput | +0.00 | [-0.03, +0.04] | 1 | |
| ➖ | dsd_uds_1mb_3k_contexts | ingress throughput | +0.00 | [-0.00, +0.01] | 1 | |
| ➖ | dsd_uds_512kb_3k_contexts | ingress throughput | +0.00 | [-0.01, +0.01] | 1 | |
| ➖ | dsd_uds_10mb_3k_contexts | ingress throughput | +0.00 | [-0.04, +0.04] | 1 | |
| ➖ | dsd_uds_1mb_50k_contexts | ingress throughput | +0.00 | [-0.00, +0.00] | 1 | |
| ➖ | dsd_uds_100mb_250k_contexts | ingress throughput | -0.00 | [-0.03, +0.02] | 1 | |
| ➖ | dsd_uds_1mb_3k_contexts_dualship | ingress throughput | -0.01 | [-0.01, +0.00] | 1 | |
| ➖ | quality_gates_idle_rss | memory utilization | -0.02 | [-0.05, +0.01] | 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
|
| @@ -243,7 +243,18 @@ where | |||
| } | |||
| } | |||
|
|
|||
| impl<'a> Tagged for &'a [&'static str] { | |||
| impl<'a> Tagged for &'a [&'a str] { | |||
There was a problem hiding this comment.
nit: It feels like a lot of these could be covered with something like
impl<T, S> Tagged for T
where
T: IntoIterator<Item=S>,
S: AsRef<str>,
{
fn visit_tags<F>(self, mut visitor: F)
where
F: FnMut(&str),
{
for tag in self {
visitor(tag.as_ref());
}
}
}
Context
In #415, we made a number of changes related to origin enrichment, one of which involved bundling the "origin" key -- a spiritual equivalent to
ContextKey, but for origin data -- intoContextKeyitself.This had the side effect of changing the size of
ContextKeyfrom 8 bytes (u64) to 24 bytes (u64+Option<OriginKey>, whereOriginKeywas just au64). AsContextKeyis the key type for our actual context cache inContextResolver, this means we're spending more memory for every cached context.Solution
We've slightly reworked how we generate, and pass around, the
OriginKeyduring context resolving. We've also folded in the hash of the origin key to the context's hash itself, rather than carrying the origin key as a separate field. This bringsContextKeyback down to 8 bytes.