T3 Code · Performance over time

Writer lock wait p95

All metrics
62 / 120
Writer lock wait p95 (lower is better)
493ms
18 Nov+2399.5% vs the median of up to 7 previous complete nightlies, regressed beyond noiseFork perf.9 5.53 ms, −73.0% vs its upstream base
AgentsRange

When upstream merges a fix, the lines meet.

Upstream nightlyPerformance forkNoise bandRange of attempts green = better, red = worse, beyond the noise band

As agents grow.

95th percentile of the time a write transaction waited for the writer connection.

Lower is better. Noise band 106.2% on Linux x64 (n = 6), so a change counts beyond 111.2%.

Unit
ms
Better
lower
Noise band
106.2%, threshold 111.2% (n = 6)
Recorded from
17 Oct 2026; before that the series is a gap, never a zero
Platforms
Linux x64
Lane
Every record, inside the agent ladder
Upstream builds
Every upstream build
Source
result.json: metrics.sql["sql.transaction|writer"].p95_lock_wait_ms

Writer lock wait p95 (lower is better) at 1 to 250 agents, Linux x64

against
Records, 50 agents
DateTagMedianMin to maxValidvs up to 7 previousVerdict
18 Nov 2026v0.1.11-nightly.20261118.4903493 ms470 to 4973+2399.5%regressed beyond noise
17 Nov 2026v0.1.11-nightly.20261117.485122.5 ms21.4 to 22.63+19.4%within noise
16 Nov 2026v0.1.10-nightly.20261116.480022.8 ms17.1 to 22.93+140.8%regressed beyond noise
15 Nov 2026v0.1.10-nightly.20261115.474919.7 ms18.8 to 22.33+51805.3%regressed beyond noise
14 Nov 2026v0.1.10-nightly.20261114.470918.9 ms18.7 to 21.53+51595.9%regressed beyond noise
13 Nov 2026v0.1.10-nightly.20261113.465621.7 ms20.2 to 23.13+61971.4%regressed beyond noise
12 Nov 2026v0.1.10-nightly.20261112.46060.025 ms0.024 to 0.0273−28.6%within noise
11 Nov 2026v0.1.10-nightly.20261111.45440.026 ms0.022 to 0.0293−25.7%within noise
10 Nov 2026v0.1.9-nightly.20261110.44910.028 ms0.027 to 0.0313−20.0%within noise
9 Nov 2026v0.1.9-nightly.20261109.44430.027 ms0.023 to 0.0293−22.9%within noise
8 Nov 2026v0.1.8-nightly.20261108.43800.024 ms0.022 to 0.0243−31.4%within noise
7 Nov 2026v0.1.8-nightly.20261107.43400.024 ms0.022 to 0.0253−31.4%within noise
6 Nov 2026v0.1.8-nightly.20261106.42880.024 ms0.023 to 0.0263−31.4%within noise
5 Nov 2026v0.1.8-nightly.20261105.42340.033 ms0.030 to 0.0373−13.2%within noise, thin reference
4 Nov 2026v0.1.8-nightly.20261104.41920.038 ms0.037 to 0.0383+8.6%within noise, thin reference
3 Nov 2026v0.1.7-nightly.20261103.41380.035 ms0.031 to 0.0353no reference yet
2 Nov 2026v0.1.7-nightly.20261102.40930.020 ms0.020 to 0.0203no reference yet
1 Nov 2026v0.1.7-nightly.20261101.40480.019 ms0.018 to 0.0193no reference yet
31 Oct 2026v0.1.7-nightly.20261031.40040.019 ms0.019 to 0.0193no reference yet
30 Oct 2026v0.1.6-nightly.20261030.39400.020 ms0.020 to 0.0203no reference yet
29 Oct 2026v0.1.6-nightly.20261029.38920.018 ms0.017 to 0.0183no reference yet
28 Oct 2026v0.1.4-nightly.20261028.38410.018 ms0.018 to 0.0183no reference yet
27 Oct 2026v0.1.4-nightly.20261027.37960.019 ms0.018 to 0.0193no reference yet
26 Oct 2026v0.1.3-nightly.20261026.37400.019 ms0.018 to 0.0193no reference yet
25 Oct 2026v0.1.3-nightly.20261025.36820.016 ms0.016 to 0.0173no reference yet
24 Oct 2026v0.1.3-nightly.20261024.36430.017 ms0.017 to 0.0173no reference yet
23 Oct 2026v0.1.3-nightly.20261023.35920.020 ms0.020 to 0.0223no reference yet
22 Oct 2026v0.1.2-nightly.20261022.354123.4 ms22.3 to 25.63no reference yet
21 Oct 2026v0.1.2-nightly.20261021.349919.4 ms18.8 to 19.43no reference yet
20 Oct 2026v0.1.1-nightly.20261020.343816.6 ms8.12 to 17.73no reference yet
19 Oct 2026v0.1.1-nightly.20261019.337313.2 ms12.2 to 19.13no noise band yet, no verdict
18 Oct 2026v0.1.1-nightly.20261018.33358.08 ms0.014 to 11.03no noise band yet, no verdict
17 Oct 2026v0.1.1-nightly.20261017.32720.003 ms0.003 to 0.0053no noise band yet, no verdict
Ladder as a table
AgentsNightly 18 NovFork perf.9
10.067 ms0.053 ms
50493 ms5.53 ms
100451 ms6.14 ms
150401 ms5.75 ms
200341 ms6 ms
250385 ms6.28 ms
Download this series