perf(engine): always populate execution cache on miss#23389

Draft

shekhirin wants to merge 5 commits intomainfrom

alexey/execution-cache-always-populate

Member

shekhirin commented Apr 8, 2026

No description provided.


          perf(engine): always populate execution cache on miss

1b1f750

github-project-automation bot moved this to Backlog in Reth Tracker

github-project-automation bot added this to Reth Tracker

Member Author

shekhirin commented Apr 8, 2026

derek bench

Member

decofe commented Apr 8, 2026 •

edited

Loading

✅ Benchmark complete! View job

Benchmark Results

⚠️ Feature is 6 commits behind main. Consider rebasing for accurate results.

Metric	`main`	`alexey/execution-cache-always-populate`	Change
Mean	32.85ms	33.73ms	+2.68% ❌ (±2.54%)
StdDev	23.80ms	23.90ms
P50	27.66ms	28.50ms	+3.03% ❌ (±2.13%)
P90	49.72ms	52.74ms	+6.08% ❌ (±5.15%)
P99	155.27ms	134.18ms	-13.59% ⚪ (±27.05%)
Mgas/s	1015.19	993.71	-2.12% ❌ (±1.18%)
Wall Clock	33.87s	34.82s	+2.80% ❌ (±2.51%)

500 blocks, 100 warmup

Wait Time Breakdown

Persistence Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	28.29ms	27.28ms
P50	0.03ms	0.03ms
P95	264.32ms	268.24ms

Trie Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.33ms	0.36ms
P50	0.00ms	0.00ms
P95	1.71ms	1.78ms

Execution Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.00ms	0.00ms
P50	0.00ms	0.00ms
P95	0.00ms	0.00ms

Grafana Dashboard

Charts

Latency, Throughput & Diff

Wait Time Breakdown

Gas vs Latency

Grafana Dashboard

View real-time metrics

Member Author

shekhirin commented Apr 8, 2026

derek bench warmup=500 blocks=2000

Member Author

shekhirin commented Apr 8, 2026

derek bench blocks=big

Member

decofe commented Apr 8, 2026 •

edited

Loading

❌ Benchmark failed while downloading snapshot. View logs

Member

decofe commented Apr 8, 2026

❌ Invalid bench command

Invalid value(s): blocks=big (must be a positive integer)

Usage: @decofe bench [blocks=N] [big-blocks] [warmup=N] [baseline=REF] [feature=REF] [samply] [no-slack] [cores=N] [abba=true|false] [otlp=true|false] [wait-time=DURATION] [baseline-args="..."] [feature-args="..."]

Member Author

shekhirin commented Apr 8, 2026

derek bench big-blocks

Member

decofe commented Apr 8, 2026 •

edited

Loading

✅ Benchmark complete! View job

Benchmark Results

⚠️ Feature is 6 commits behind main. Consider rebasing for accurate results.

Metric	`main`	`alexey/execution-cache-always-populate`	Change
Mean	715.77ms	708.63ms	-1.00% ⚪ (±1.22%)
StdDev	149.43ms	146.87ms
P50	685.96ms	678.89ms	-1.03% ⚪ (±1.58%)
P90	901.27ms	890.35ms	-1.21% ⚪ (±3.09%)
P99	1242.35ms	1255.97ms	+1.10% ⚪ (±9.46%)
Mgas/s	1477.79	1490.97	+0.89% ⚪ (±1.09%)
Wall Clock	745.59s	738.07s	-1.01% ⚪ (±1.58%)

500 big blocks, 100 warmup

Wait Time Breakdown

Persistence Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	1.69ms	1.84ms
P50	0.08ms	0.08ms
P95	0.11ms	0.11ms

Trie Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.02ms	0.00ms
P50	0.00ms	0.00ms
P95	0.00ms	0.00ms

Execution Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.00ms	0.00ms
P50	0.00ms	0.00ms
P95	0.00ms	0.00ms

Grafana Dashboard

Charts

Latency, Throughput & Diff

Wait Time Breakdown

Gas vs Latency

Grafana Dashboard

View real-time metrics

Member Author

shekhirin commented Apr 8, 2026

derek bench warmup=500 blocks=2000

Member

decofe commented Apr 8, 2026 •

edited

Loading

❌ Benchmark failed while downloading snapshot. View logs

Member Author

shekhirin commented Apr 8, 2026

derek bench warmup=500 blocks=2000

Member

decofe commented Apr 8, 2026 •

edited

Loading

❌ Benchmark failed while downloading snapshot. View logs


          Merge branch 'main' into alexey/execution-cache-always-populate

50ee6fc

Member Author

shekhirin commented Apr 8, 2026

derek bench warmup=500 blocks=2000

Member

decofe commented Apr 8, 2026 •

edited

Loading

✅ Benchmark complete! View job

Benchmark Results

Metric	`main`	`alexey/execution-cache-always-populate`	Change
Mean	28.29ms	28.55ms	+0.93% ❌ (±0.90%)
StdDev	14.64ms	14.51ms
P50	25.80ms	26.10ms	+1.14% ⚪ (±1.33%)
P90	44.69ms	45.35ms	+1.49% ⚪ (±1.78%)
P99	79.30ms	78.07ms	-1.55% ⚪ (±9.79%)
Mgas/s	1087.92	1078.00	-0.91% ❌ (±0.53%)
Wall Clock	117.05s	118.00s	+0.81% ⚪ (±0.88%)

2000 blocks, 500 warmup

Wait Time Breakdown

Persistence Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	27.51ms	27.34ms
P50	0.03ms	0.03ms
P95	248.07ms	241.36ms

Trie Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.33ms	0.33ms
P50	0.00ms	0.00ms
P95	1.57ms	1.57ms

Execution Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.00ms	0.00ms
P50	0.00ms	0.00ms
P95	0.00ms	0.00ms

Grafana Dashboard

Charts

Latency, Throughput & Diff

Wait Time Breakdown

Gas vs Latency

Grafana Dashboard

View real-time metrics

Member Author

shekhirin commented Apr 8, 2026

derek bench warmup=500 blocks=2000 samply

Member

decofe commented Apr 8, 2026 •

edited

Loading

✅ Benchmark complete! View job

Benchmark Results

⚠️ Feature is 1 commit behind main. Consider rebasing for accurate results.

Metric	`main`	`alexey/execution-cache-always-populate`	Change
Mean	31.01ms	31.35ms	+1.07% ❌ (±0.71%)
StdDev	14.92ms	15.08ms
P50	28.57ms	28.86ms	+1.04% ⚪ (±1.23%)
P90	48.63ms	49.45ms	+1.68% ❌ (±1.55%)
P99	80.48ms	81.84ms	+1.69% ⚪ (±7.59%)
Mgas/s	980.84	971.16	-0.99% ❌ (±0.48%)
Wall Clock	128.39s	129.61s	+0.96% ❌ (±0.69%)

2000 blocks, 500 warmup

Wait Time Breakdown

Persistence Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	28.10ms	27.68ms
P50	0.03ms	0.03ms
P95	243.39ms	242.57ms

Trie Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.38ms	0.39ms
P50	0.00ms	0.00ms
P95	1.81ms	1.90ms

Execution Cache Update Wait

Metric	`main`	`alexey/execution-cache-always-populate`
Mean	0.00ms	0.00ms
P50	0.00ms	0.00ms
P95	0.00ms	0.00ms

Grafana Dashboard

Charts

Latency, Throughput & Diff

Wait Time Breakdown

Gas vs Latency

Grafana Dashboard

View real-time metrics

shekhirin and others added 2 commits

April 9, 2026 20:02


          Merge branch 'main' into alexey/execution-cache-always-populate

1fbde32


          fix: remove unused metrics field from CachedStateProvider

bd18b63

Co-authored-by: Alexey Shekhirin <5773434+shekhirin@users.noreply.github.com>
Amp-Thread-ID: https://ampcode.com/threads/T-019d73a0-5242-7629-b741-60d124223bda

Member Author

shekhirin commented Apr 10, 2026

derek bench samply

Member

decofe commented Apr 10, 2026 •

edited

Loading

❌ Benchmark failed while building binaries. View logs


          fix: remove unused metrics arg from CachedStateProvider::new in paylo…

dc272d8

…ad builder

Co-Authored-By: Alexey Shekhirin <5773434+shekhirin@users.noreply.github.com>
Amp-Thread-ID: https://ampcode.com/threads/T-019d76a9-dc7b-710c-9c49-9ce39c07e98f

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

Reviewers

mattsse Awaiting requested review from mattsse mattsse will be requested when the pull request is marked ready for review mattsse is a code owner

Rjected Awaiting requested review from Rjected Rjected will be requested when the pull request is marked ready for review Rjected is a code owner

mediocregopher Awaiting requested review from mediocregopher mediocregopher will be requested when the pull request is marked ready for review mediocregopher is a code owner

At least 1 approving review is required to merge this pull request.

Labels

None yet