Skip to content

perf: use cached prewarm fee token#5558

Closed
tempoxyz-bot wants to merge 2 commits into
mainfrom
auto-opt/20260610-cache-prewarm-fee-token
Closed

perf: use cached prewarm fee token#5558
tempoxyz-bot wants to merge 2 commits into
mainfrom
auto-opt/20260610-cache-prewarm-fee-token

Conversation

@tempoxyz-bot

@tempoxyz-bot tempoxyz-bot commented Jun 10, 2026

Copy link
Copy Markdown
Contributor

Use the pooled transaction's effective fee token helper when collecting prewarm storage touches. This reuses validation-cached fee-token resolution instead of rebuilding the same fallback in the payload prewarmer.

perf: use cached prewarm fee token
0d1f972 
Use the pooled transaction's effective fee token helper when collecting prewarm storage touches. This reuses validation-cached fee-token resolution instead of rebuilding the same fallback in the payload prewarmer.
@tempoxyz-bot tempoxyz-bot self-assigned this Jun 10, 2026
@tempoxyz-bot

tempoxyz-bot commented Jun 10, 2026

Copy link
Copy Markdown
Contributor Author

derek bench mode=e2e otlp=false

@decofe

decofe commented Jun 10, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

cc @tempoxyz-bot

✅ Benchmark complete: Improvement View job

✅ Bench Comparison: Improvement

Refs: main vs auto-opt/20260610-cache-prewarm-fee-token
Criteria: 95% run-bootstrap CI must clear floor; cells show delta (+/-CI/floor).

Configuration

  • Derek command: derek bench mode=e2e preset=tip20 duration=90 bloat=100 tps=50000 accounts=1000 max-concurrent-requests=500 baseline=main feature=auto-opt/20260610-cache-prewarm-fee-token baseline-hardfork="" feature-hardfork="" gas-limit=1000000000 run-pairs=3 otlp=false metrics=false no-cache=false force-bloat=false
  • Bloat: 100000 MiB
  • Preset: tip20
  • Target TPS: 50000
  • Duration: 90s
  • Run pairs: 3
  • Baseline blocks: 477
  • Feature blocks: 478

Tempo Metrics

Metric Baseline Feature Delta
TPS Mean 21960 21949 -0.05% ⚪ (+/-0.51/floor 0.55)
Gas Throughput [Mgas/s] 971.5 971.1 -0.04% ⚪ (+/-0.51/floor 0.50)
Block Time Mean [ms] 558.8 557.6 -0.21% ⚪ (+/-0.11/floor 0.40)
Block Time P50 [ms] 559.0 559.0 +0.00% ⚪ (+/-0.48/floor 0.70)
Block Time P90 [ms] 591.0 589.0 -0.34% ⚪ (+/-0.79/floor 0.70)
Block Time P99 [ms] 621.0 632.0 +1.77% ⚪ (+/-1.21/floor 1.60)
Serialized Block Size / Tx P50 [B/tx] 251.1 251.1 +0.00% ⚪ (+/-0.00/floor 0.70)
Serialized Block Size / Tx P90 [B/tx] 251.1 251.1 +0.00% ⚪ (+/-5.20/floor 0.70)
Serialized Block Size / Tx P99 [B/tx] 305.2 251.1 -17.73% ✅ (+/-13.94/floor 0.70)

Builder

Metric Baseline Feature Delta
Gas Throughput [Mgas/s] 2335.8 2332.1 -0.16% ⚪ (+/-0.69/floor 0.70)
P50 [ms] 230.0 229.6 -0.17% ⚪ (+/-0.24/floor 0.35)
P90 [ms] 243.7 243.5 -0.08% ⚪ (+/-0.64/floor 0.70)
P99 [ms] 274.0 279.7 +2.08% ⚪ (+/-2.12/floor 0.95)
Builder details
Metric Baseline Feature Delta
Finish P50 [ms] 23.0 23.5 +2.17%
Finish P90 [ms] 33.5 33.2 -0.90%
Finish P99 [ms] 39.3 39.5 +0.51%
Pool Fetch P50 [ms] 9.0 8.4 -6.67%
Pool Fetch P90 [ms] 19.5 20.0 +2.56%
Pool Fetch P99 [ms] 24.4 25.6 +4.92%
Included Tx Exec P50 [ms] - -
Included Tx Exec P90 [ms] - -
Included Tx Exec P99 [ms] - -
Invalid Tx Exec P50 [ms] - -
Invalid Tx Exec P90 [ms] - -
Invalid Tx Exec P99 [ms] - -
Invalid Tx Attempts P50 0.0 0.0 0.00%
Invalid Tx Attempts P90 0.0 0.0 0.00%
Invalid Tx Attempts P99 0.0 0.0 0.00%
Invalid Tx Skips 0 0 0.00%
Nonce Too Low Skips 0 0 0.00%
Serialized Block Size P50 [KiB] 3007.0 3002.6 -0.15%
Serialized Block Size P90 [KiB] 3348.7 3361.2 +0.37%
Serialized Block Size P99 [KiB] 3714.4 3724.7 +0.28%
Fill Overhead P50 [ms] - -
Fill Overhead P90 [ms] - -
Fill Overhead P99 [ms] - -
Fill Idle P50 [ms] 0.0 0.0 0.00%
Fill Idle P90 [ms] 30.0 26.0 -13.33%
Fill Idle P99 [ms] 94.0 107.0 +13.83%

Validator

Metric Baseline Feature Delta
Gas Throughput [Mgas/s] 2262.9 2271.3 +0.37% ⚪ (+/-0.63/floor 0.65)
P50 [ms] 241.2 241.0 -0.08% ⚪ (+/-0.64/floor 1.55)
P90 [ms] 266.2 265.1 -0.41% ⚪ (+/-1.23/floor 1.55)
P99 [ms] 286.9 281.1 -2.02% ⚪ (+/-1.86/floor 2.05)

Observability

@tempoxyz-bot tempoxyz-bot marked this pull request as ready for review June 10, 2026 02:19
@tempoxyz-bot

tempoxyz-bot commented Jun 10, 2026

Copy link
Copy Markdown
Contributor Author

derek bench samply run-pairs=1 duration=300 no-slack

@decofe

decofe commented Jun 10, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

cc @tempoxyz-bot

⚪ Benchmark complete: No Difference View job

⚪ Bench Comparison: No Difference

Refs: main vs auto-opt/20260610-cache-prewarm-fee-token
Criteria: 95% run-bootstrap CI must clear floor; cells show delta (+/-CI/floor).

Configuration

  • Derek command: derek bench mode=e2e preset=tip20 duration=300 bloat=100 tps=50000 accounts=1000 max-concurrent-requests=500 baseline=main feature=auto-opt/20260610-cache-prewarm-fee-token baseline-hardfork="" feature-hardfork="" gas-limit=1000000000 run-pairs=1 otlp=true metrics=false no-cache=false force-bloat=false samply
  • Bloat: 100000 MiB
  • Preset: tip20
  • Target TPS: 50000
  • Duration: 300s
  • Run pairs: 1
  • Baseline blocks: 545
  • Feature blocks: 545

Tempo Metrics

Metric Baseline Feature Delta
TPS Mean 21322 21093 -1.07% ⚪
Gas Throughput [Mgas/s] 943.3 933.2 -1.07% ⚪
Block Time Mean [ms] 548.5 548.3 -0.04% ⚪
Block Time P50 [ms] 549.0 548.0 -0.18% ⚪
Block Time P90 [ms] 588.0 591.0 +0.51% ⚪
Block Time P99 [ms] 622.0 629.0 +1.13% ⚪
Serialized Block Size / Tx P50 [B/tx] 251.1 251.1 +0.00% ⚪
Serialized Block Size / Tx P90 [B/tx] 251.1 251.1 +0.00% ⚪
Serialized Block Size / Tx P99 [B/tx] 251.1 251.1 +0.00% ⚪

Builder

Metric Baseline Feature Delta
Gas Throughput [Mgas/s] 2250.1 2230.2 -0.88% ⚪
P50 [ms] 229.2 228.2 -0.44% ⚪
P90 [ms] 247.0 245.7 -0.53% ⚪
P99 [ms] 273.2 277.6 +1.61% ⚪
Builder details
Metric Baseline Feature Delta
Finish P50 [ms] 29.9 30.3 +1.34%
Finish P90 [ms] 46.2 47.2 +2.16%
Finish P99 [ms] 57.6 56.3 -2.26%
Pool Fetch P50 [ms] 10.1 10.4 +2.97%
Pool Fetch P90 [ms] 25.1 23.6 -5.98%
Pool Fetch P99 [ms] 34.2 32.7 -4.39%
Included Tx Exec P50 [ms] - -
Included Tx Exec P90 [ms] - -
Included Tx Exec P99 [ms] - -
Invalid Tx Exec P50 [ms] - -
Invalid Tx Exec P90 [ms] - -
Invalid Tx Exec P99 [ms] - -
Invalid Tx Attempts P50 0.0 0.0 0.00%
Invalid Tx Attempts P90 0.0 0.0 0.00%
Invalid Tx Attempts P99 0.0 0.0 0.00%
Invalid Tx Skips 0 0 0.00%
Nonce Too Low Skips 0 0 0.00%
Serialized Block Size P50 [KiB] 2877.6 2828.8 -1.70%
Serialized Block Size P90 [KiB] 3291.9 3246.5 -1.38%
Serialized Block Size P99 [KiB] 3792.4 3744.6 -1.26%
Fill Overhead P50 [ms] - -
Fill Overhead P90 [ms] - -
Fill Overhead P99 [ms] - -
Fill Idle P50 [ms] 0.0 0.0 0.00%
Fill Idle P90 [ms] 54.0 50.0 -7.41%
Fill Idle P99 [ms] 120.0 124.0 +3.33%

Validator

Metric Baseline Feature Delta
Gas Throughput [Mgas/s] 2298.5 2275.0 -1.02% ⚪
P50 [ms] 228.1 227.1 -0.44% ⚪
P90 [ms] 258.2 256.3 -0.74% ⚪
P99 [ms] 280.5 281.2 +0.25% ⚪

Observability

Samply Profiles

@mediocregopher

mediocregopher commented Jun 12, 2026

Copy link
Copy Markdown
Contributor

derek bench duration=120

@decofe

decofe commented Jun 12, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

cc @mediocregopher

❌ Benchmark complete: Regression View job

❌ Bench Comparison: Regression

Refs: main vs auto-opt/20260610-cache-prewarm-fee-token
Criteria: 95% run-bootstrap CI must clear floor; cells show delta (+/-CI/floor).

Configuration

  • Derek command: derek bench mode=e2e preset=tip20 duration=120 bloat=100 token-count=4 tps=50000 accounts=1000 max-concurrent-requests=500 baseline=main feature=auto-opt/20260610-cache-prewarm-fee-token baseline-hardfork="" feature-hardfork="" gas-limit=1000000000 run-pairs=3 otlp=true metrics=false no-cache=false force-bloat=false
  • Bloat: 100000 MiB
  • Token count: 4
  • Preset: tip20
  • Target TPS: 50000
  • Duration: 120s
  • Run pairs: 3
  • Baseline blocks: 661
  • Feature blocks: 666

Tempo Metrics

Metric Baseline Feature Delta
TPS Mean 19006 18823 -0.96% ❌ (+/-0.31/floor 0.55)
Gas Throughput [Mgas/s] 962.4 953.1 -0.97% ❌ (+/-0.32/floor 0.50)
Block Time Mean [ms] 541.2 538.8 -0.44% ⚪ (+/-0.25/floor 0.40)
Block Time P50 [ms] 540.0 538.0 -0.37% ⚪ (+/-0.93/floor 0.70)
Block Time P90 [ms] 583.0 583.0 +0.00% ⚪ (+/-0.69/floor 0.70)
Block Time P99 [ms] 634.0 625.0 -1.42% ⚪ (+/-2.60/floor 1.60)
Serialized Block Size / Tx P50 [B/tx] 251.1 251.1 +0.00% ⚪ (+/-0.00/floor 0.70)
Serialized Block Size / Tx P90 [B/tx] 251.1 251.1 +0.00% ⚪ (+/-0.00/floor 0.70)
Serialized Block Size / Tx P99 [B/tx] 251.1 251.1 +0.00% ⚪ (+/-0.00/floor 0.70)

Builder

Metric Baseline Feature Delta
Gas Throughput [Mgas/s] 2260.4 2227.0 -1.48% ⚪ (+/-0.56/floor 0.95)
P50 [ms] 228.7 229.5 +0.35% ⚪ (+/-0.37/floor 0.45)
P90 [ms] 244.1 243.8 -0.12% ⚪ (+/-0.55/floor 0.90)
P99 [ms] 271.7 276.5 +1.77% ⚪ (+/-3.90/floor 1.25)
Builder details
Metric Baseline Feature Delta
Finish P50 [ms] 41.1 42.0 +2.19%
Finish P90 [ms] 54.8 55.5 +1.28%
Finish P99 [ms] 61.6 62.5 +1.46%
Pool Fetch P50 [ms] 7.3 8.0 +9.59%
Pool Fetch P90 [ms] 17.6 18.4 +4.55%
Pool Fetch P99 [ms] 24.4 23.9 -2.05%
Included Tx Exec P50 [ms] - -
Included Tx Exec P90 [ms] - -
Included Tx Exec P99 [ms] - -
Invalid Tx Exec P50 [ms] - -
Invalid Tx Exec P90 [ms] - -
Invalid Tx Exec P99 [ms] - -
Invalid Tx Attempts P50 0.0 0.0 0.00%
Invalid Tx Attempts P90 0.0 0.0 0.00%
Invalid Tx Attempts P99 0.0 0.0 0.00%
Invalid Tx Skips 0 0 0.00%
Nonce Too Low Skips 0 0 0.00%
Serialized Block Size P50 [KiB] 2477.8 2436.6 -1.66%
Serialized Block Size P90 [KiB] 2881.5 2855.3 -0.91%
Serialized Block Size P99 [KiB] 3514.4 3422.7 -2.61%
Fill Overhead P50 [ms] - -
Fill Overhead P90 [ms] - -
Fill Overhead P99 [ms] - -
Fill Idle P50 [ms] 0.0 0.0 0.00%
Fill Idle P90 [ms] 0.0 0.0 0.00%
Fill Idle P99 [ms] 86.0 90.0 +4.65%

Validator

Metric Baseline Feature Delta
Gas Throughput [Mgas/s] 2237.7 2227.4 -0.46% ⚪ (+/-0.53/floor 0.65)
P50 [ms] 232.1 230.5 -0.69% ⚪ (+/-1.11/floor 1.55)
P90 [ms] 264.4 265.7 +0.49% ⚪ (+/-1.10/floor 1.55)
P99 [ms] 304.5 292.5 -3.94% ⚪ (+/-3.33/floor 2.05)

Observability

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@0xKitsune 0xKitsune Awaiting requested review from 0xKitsune 0xKitsune is a code owner

@klkvr klkvr Awaiting requested review from klkvr klkvr is a code owner

@mattsse mattsse Awaiting requested review from mattsse mattsse is a code owner

Assignees

@tempoxyz-bot tempoxyz-bot

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@tempoxyz-bot @decofe @mediocregopher