Conversation
9d7dace to
c17b1eb
Compare
BenchmarksStartupParameters
See matching parameters
SummaryFound 0 performance improvements and 3 performance regressions! Performance is the same for 42 metrics, 9 unstable metrics.
LoadParameters
See matching parameters
SummaryFound 0 performance improvements and 0 performance regressions! Performance is the same for 11 metrics, 15 unstable metrics. Request duration reports for insecure-bankgantt
title insecure-bank - request duration [CI 0.99] : candidate=1.31.0-SNAPSHOT~db1129529b, baseline=1.31.0-SNAPSHOT~aee3ca59c6
dateFormat X
axisFormat %s
section baseline
no_agent (367.04 µs) : 347, 387
. : milestone, 367,
iast (477.188 µs) : 457, 498
. : milestone, 477,
iast_FULL (532.938 µs) : 513, 553
. : milestone, 533,
iast_GLOBAL (499.922 µs) : 480, 520
. : milestone, 500,
iast_HARDCODED_SECRET_DISABLED (465.464 µs) : 445, 486
. : milestone, 465,
iast_INACTIVE (450.4 µs) : 429, 472
. : milestone, 450,
iast_TELEMETRY_OFF (473.792 µs) : 453, 495
. : milestone, 474,
tracing (452.805 µs) : 432, 474
. : milestone, 453,
section candidate
no_agent (360.342 µs) : 341, 380
. : milestone, 360,
iast (472.448 µs) : 452, 493
. : milestone, 472,
iast_FULL (538.814 µs) : 518, 559
. : milestone, 539,
iast_GLOBAL (497.757 µs) : 476, 520
. : milestone, 498,
iast_HARDCODED_SECRET_DISABLED (468.158 µs) : 447, 489
. : milestone, 468,
iast_INACTIVE (440.098 µs) : 420, 460
. : milestone, 440,
iast_TELEMETRY_OFF (471.325 µs) : 450, 492
. : milestone, 471,
tracing (447.451 µs) : 427, 468
. : milestone, 447,
Request duration reports for petclinicgantt
title petclinic - request duration [CI 0.99] : candidate=1.31.0-SNAPSHOT~db1129529b, baseline=1.31.0-SNAPSHOT~aee3ca59c6
dateFormat X
axisFormat %s
section baseline
no_agent (1.345 ms) : 1326, 1364
. : milestone, 1345,
appsec (1.774 ms) : 1750, 1798
. : milestone, 1774,
iast (1.537 ms) : 1513, 1560
. : milestone, 1537,
profiling (1.535 ms) : 1511, 1560
. : milestone, 1535,
tracing (1.517 ms) : 1493, 1541
. : milestone, 1517,
section candidate
no_agent (1.338 ms) : 1319, 1357
. : milestone, 1338,
appsec (1.77 ms) : 1746, 1795
. : milestone, 1770,
iast (1.539 ms) : 1515, 1562
. : milestone, 1539,
profiling (1.512 ms) : 1487, 1536
. : milestone, 1512,
tracing (1.509 ms) : 1486, 1533
. : milestone, 1509,
|
40143a2 to
65a03e7
Compare
…ded before classes that extend/implement them.
65a03e7 to
db11295
Compare
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
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 Do
Updates
j.l.ClassLoaderto notify the tracer before defining a class. The tracer can then attempt to preload the superclass and any interfaces, which will make sure those types are memoized/cached before the initial type is matched.To disable this optimization add this JVM option:
or set this environment variable:
Motivation
For spring-petclinic (when memoization is enabled) this reduces the number of class-file lookups by over a third.
(from 1970 lookups to 1226 lookups)
Note on Benchmarking Data
Benchmarks indicate that there is a minor startup regression in this PR, but this is untrue because the Datadog benchmarking platform measures the time between phases of tracer installation, not application startup as a whole.
What actually happened here is we are now instrumenting an additional method from j.l.ClassLoader, which incurs a non-zero cost, but it has the benefit of optimizing how we use various type related caches. (Specifically around re-using memoized results from the type hierarchy.) This means that instrumented applications as a whole will startup faster, but they might spend a few milliseconds extra installing the agent. Those additional milliseconds are more than recouped over the life of the application by savings when matching application types.
Jira ticket: APMS-11443