Skip blanking a page that was never used

[martian56]

Retest under the paired harness. This was measured once before it existed and the numbers were unusable: arm64 plus 10.9, 1cpu minus 17.0, x64-full minus 88.7 with write p99 at 115 ms, which is a degraded runner rather than a result.

Rolling onto a new packing page blanks it through a full commit round trip, and that happens while the open page lock is held, so every other writer waits on it. Only a page taken off the free pool needs blanking. A page that has never been used is past the end of the file and already reads as empty, so the round trip is pure cost.

Now that both builds run on the same runner this can be judged.

[coderabbitai]

Warning

Review limit reached

@martian56, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 19 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: f4cad7a8-6dda-4e1e-9b0f-5291fbce3663

📥 Commits

Reviewing files that changed from the base of the PR and between b70d500 and 542c84f.

📒 Files selected for processing (3)

• mvcc/src/database.rs
• storage/src/group_commit.rs
• storage/src/store.rs

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch skip-clear-retest

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

Bench report

✅ Best improvement: windows-x64-full mixed at +34.4%. Nothing regressed beyond noise.

Baseline: main @ same runn. Cells read baseline → this PR. Verdicts need more than ±15% to leave the noise band.

linux-arm64-full

workload	ops/s	Δ ops/s	read p99 ms	write p99 ms	write p99.9 ms
reads	32956 → 32586	⚪ -1.1%	0.21 → 0.22	-	-
mixed	14889 → 13681	⚪ -8.1%	0.43 → 0.43	13.06 → 19.06	33.01 → 34.25
writes	6225 → 6038	⚪ -3.0%	-	6.03 → 6.22	30.08 → 29.52

linux-x64-1cpu-1gb

workload	ops/s	Δ ops/s	read p99 ms	write p99 ms	write p99.9 ms
reads	22618	new	0.23	-	-
mixed	5377	new	0.38	44.83	202.89
writes	2483	new	-	20.43	169.99

linux-x64-2cpu-2gb

workload	ops/s	Δ ops/s	read p99 ms	write p99 ms	write p99.9 ms
reads	15003	new	0.42	-	-
mixed	10263	new	0.57	4.93	18.18
writes	5236	new	-	2.38	11.83

linux-x64-full

workload	ops/s	Δ ops/s	read p99 ms	write p99 ms	write p99.9 ms
reads	24760 → 24959	⚪ +0.8%	0.28 → 0.27	-	-
mixed	14371 → 14388	⚪ +0.1%	0.43 → 0.38	4.21 → 4.13	17.91 → 18.59
writes	6133 → 6287	⚪ +2.5%	-	2.13 → 2.11	11.84 → 11.87

windows-x64-full

workload	ops/s	Δ ops/s	read p99 ms	write p99 ms	write p99.9 ms
reads	22470 → 21299	⚪ -5.2%	0.37 → 0.50	-	-
mixed	5413 → 7274	🟢 +34.4%	0.35 → 0.34	30.61 → 22.35	41.27 → 34.55
writes	826 → 755	⚪ -8.6%	-	16.29 → 16.41	27.89 → 33.50

Field reference

All engines in identical containers, same driver, measured 2026-07-24 @ ba0ec39. A different harness than the tables above, so read this as standing, not as this PR's delta.

linux-arm64-full

engine	reads	mixed	writes
quantadb	17978	12420 (27ms wp99)	6449 (19ms wp99)
postgres18	15227	14723 (2ms wp99)	9923 (1ms wp99)
mariadb114	4221	4437 (8ms wp99)	6666 (3ms wp99)
mysql84	4258	4394 (9ms wp99)	4274 (4ms wp99)

linux-x64-1cpu-1gb

engine	reads	mixed	writes
quantadb	19419	10854 (29ms wp99)	5170 (39ms wp99)
postgres18	10196	8580 (43ms wp99)	5912 (46ms wp99)
mariadb114	3455	3732 (4ms wp99)	5626 (8ms wp99)
mysql84	3497	3628 (4ms wp99)	3259 (36ms wp99)

linux-x64-2cpu-2gb

engine	reads	mixed	writes
quantadb	22528	16774 (5ms wp99)	8374 (4ms wp99)
postgres18	17060	17006 (2ms wp99)	11404 (1ms wp99)
mariadb114	3735	4006 (3ms wp99)	6741 (3ms wp99)
mysql84	3804	3845 (4ms wp99)	4847 (3ms wp99)

linux-x64-full

engine	reads	mixed	writes
quantadb	18144	15075 (6ms wp99)	8025 (5ms wp99)
postgres18	15402	15154 (2ms wp99)	10872 (1ms wp99)
mariadb114	3437	3699 (4ms wp99)	6138 (3ms wp99)
mysql84	3490	3615 (4ms wp99)	3996 (5ms wp99)

Shared CI runners: treat single-profile swings as suggestive, agreement across profiles as real.

[martian56]

Measured against a same runner baseline and rejected as neutral.

• linux-x64-full: plus 2.5 percent
• linux-arm64-full: minus 3.0 percent
• windows-x64-full: minus 8.6 percent
• docker profiles had no paired baseline this run

All inside the band. The reasoning was sound but the cost is not there: a rollover happens roughly once per fifty records, so even a full commit round trip under the lock amortises to almost nothing per write.

Worth contrasting with the first attempt, which reported plus 10.9, minus 17.0 and minus 88.7 on three profiles for this same change. That spread was the harness, not the code, and it is the clearest example yet of why the paired baseline was worth building.