From 3a30b4ac3be3df21b02de20994cf3c74418b6815 Mon Sep 17 00:00:00 2001
From: "shoney.arickathil" <shoney.john@outlook.com>
Date: Fri, 21 Aug 2026 16:54:13 +0200
Subject: [PATCH] =?UTF-8?q?feat(db-bench):=20first=20baseline=20=E2=80=94?=
 =?UTF-8?q?=20the=20measurement=20contract=20(T5)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- bench/baseline.json: 74 metrics from the first full campaign;
  tolerances tuned by a two-run repeatability check (mix* 50%,
  read/query 35%, rest 15% — rationale in _config)
- gate bites: --check mode; doctored copy fails, both real runs 74/0
- headline: durable seed 4.5k/s vs ram 297k/s (23's case); reads
  O(table) at ~1.5k/s; mixread 1280 vs 21 ops/s single-vs-multi
  (the arc's price); msgrate 13.4M vs 2.45M (mutex-inbox number)
- arc delta recorded in story 8; findings in sample README

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---
 bench/baseline.json                           | 453 ++++++++++++++++++
 docs/examples/db-bench/README.md              |  21 +
 .../done/08-shard-actor-runtime.md            |  13 +
 docs/superpowers/plans/2026-08-21-db-bench.md |  25 +-
 scripts/db-bench.py                           |   8 +
 5 files changed, 511 insertions(+), 9 deletions(-)
 create mode 100644 bench/baseline.json

diff --git a/bench/baseline.json b/bench/baseline.json
new file mode 100644
index 0000000..9273cc8
--- /dev/null
+++ b/bench/baseline.json
@@ -0,0 +1,453 @@
+{
+ "_config": {
+  "N": 20000,
+  "crash_reps": 3,
+  "msg_n": 200000,
+  "note": "refresh only with a commit that says why; tolerances widened 2026-08-21 from the two-run repeatability check: mix* 50% (scheduling-dependent small counts), read/query 35% (machine jitter), everything else 15%",
+  "wal_n": 4000
+ },
+ "durable.s1.mixread.ops_sec": {
+  "dir": "higher",
+  "floor": 187,
+  "tolerance_pct": 50,
+  "value": 751
+ },
+ "durable.s1.mixread.p50us": {
+  "dir": "lower",
+  "floor": 11844,
+  "tolerance_pct": 50,
+  "value": 2961
+ },
+ "durable.s1.mixread.p99us": {
+  "dir": "lower",
+  "floor": 19044,
+  "tolerance_pct": 50,
+  "value": 4761
+ },
+ "durable.s1.mixwrite.ops_sec": {
+  "dir": "higher",
+  "floor": 20,
+  "tolerance_pct": 50,
+  "value": 83
+ },
+ "durable.s1.mixwrite.p50us": {
+  "dir": "lower",
+  "floor": 19408,
+  "tolerance_pct": 50,
+  "value": 4852
+ },
+ "durable.s1.mixwrite.p99us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "durable.s1.query.ops_sec": {
+  "dir": "higher",
+  "floor": 335,
+  "tolerance_pct": 35,
+  "value": 1341
+ },
+ "durable.s1.query.p50us": {
+  "dir": "lower",
+  "floor": 2996,
+  "tolerance_pct": 35,
+  "value": 749
+ },
+ "durable.s1.query.p99us": {
+  "dir": "lower",
+  "floor": 3204,
+  "tolerance_pct": 35,
+  "value": 801
+ },
+ "durable.s1.read.ops_sec": {
+  "dir": "higher",
+  "floor": 336,
+  "tolerance_pct": 35,
+  "value": 1346
+ },
+ "durable.s1.read.p50us": {
+  "dir": "lower",
+  "floor": 3032,
+  "tolerance_pct": 35,
+  "value": 758
+ },
+ "durable.s1.read.p99us": {
+  "dir": "lower",
+  "floor": 3324,
+  "tolerance_pct": 35,
+  "value": 831
+ },
+ "durable.s1.seed.ops_sec": {
+  "dir": "higher",
+  "floor": 1123,
+  "tolerance_pct": 15,
+  "value": 4492
+ },
+ "durable.s1.seed.p50us": {
+  "dir": "lower",
+  "floor": 840,
+  "tolerance_pct": 15,
+  "value": 210
+ },
+ "durable.s1.seed.p99us": {
+  "dir": "lower",
+  "floor": 2188,
+  "tolerance_pct": 15,
+  "value": 547
+ },
+ "durable.s1.write.ops_sec": {
+  "dir": "higher",
+  "floor": 307,
+  "tolerance_pct": 15,
+  "value": 1230
+ },
+ "durable.s1.write.p50us": {
+  "dir": "lower",
+  "floor": 3036,
+  "tolerance_pct": 15,
+  "value": 759
+ },
+ "durable.s1.write.p99us": {
+  "dir": "lower",
+  "floor": 7232,
+  "tolerance_pct": 15,
+  "value": 1808
+ },
+ "durable.sN.mixread.ops_sec": {
+  "dir": "higher",
+  "floor": 5,
+  "tolerance_pct": 50,
+  "value": 20
+ },
+ "durable.sN.mixread.p50us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "durable.sN.mixread.p99us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "durable.sN.mixwrite.ops_sec": {
+  "dir": "higher",
+  "floor": 0,
+  "tolerance_pct": 50,
+  "value": 2
+ },
+ "durable.sN.mixwrite.p50us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "durable.sN.mixwrite.p99us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "durable.sN.query.ops_sec": {
+  "dir": "higher",
+  "floor": 328,
+  "tolerance_pct": 35,
+  "value": 1312
+ },
+ "durable.sN.query.p50us": {
+  "dir": "lower",
+  "floor": 3028,
+  "tolerance_pct": 35,
+  "value": 757
+ },
+ "durable.sN.query.p99us": {
+  "dir": "lower",
+  "floor": 3392,
+  "tolerance_pct": 35,
+  "value": 848
+ },
+ "durable.sN.read.ops_sec": {
+  "dir": "higher",
+  "floor": 350,
+  "tolerance_pct": 35,
+  "value": 1403
+ },
+ "durable.sN.read.p50us": {
+  "dir": "lower",
+  "floor": 3000,
+  "tolerance_pct": 35,
+  "value": 750
+ },
+ "durable.sN.read.p99us": {
+  "dir": "lower",
+  "floor": 3420,
+  "tolerance_pct": 35,
+  "value": 855
+ },
+ "durable.sN.seed.ops_sec": {
+  "dir": "higher",
+  "floor": 1119,
+  "tolerance_pct": 15,
+  "value": 4478
+ },
+ "durable.sN.seed.p50us": {
+  "dir": "lower",
+  "floor": 836,
+  "tolerance_pct": 15,
+  "value": 209
+ },
+ "durable.sN.seed.p99us": {
+  "dir": "lower",
+  "floor": 2372,
+  "tolerance_pct": 15,
+  "value": 593
+ },
+ "durable.sN.write.ops_sec": {
+  "dir": "higher",
+  "floor": 316,
+  "tolerance_pct": 15,
+  "value": 1265
+ },
+ "durable.sN.write.p50us": {
+  "dir": "lower",
+  "floor": 3064,
+  "tolerance_pct": 15,
+  "value": 766
+ },
+ "durable.sN.write.p99us": {
+  "dir": "lower",
+  "floor": 6616,
+  "tolerance_pct": 15,
+  "value": 1654
+ },
+ "ram.s1.mixread.ops_sec": {
+  "dir": "higher",
+  "floor": 320,
+  "tolerance_pct": 50,
+  "value": 1280
+ },
+ "ram.s1.mixread.p50us": {
+  "dir": "lower",
+  "floor": 10904,
+  "tolerance_pct": 50,
+  "value": 2726
+ },
+ "ram.s1.mixread.p99us": {
+  "dir": "lower",
+  "floor": 12472,
+  "tolerance_pct": 50,
+  "value": 3118
+ },
+ "ram.s1.mixwrite.ops_sec": {
+  "dir": "higher",
+  "floor": 35,
+  "tolerance_pct": 50,
+  "value": 142
+ },
+ "ram.s1.mixwrite.p50us": {
+  "dir": "lower",
+  "floor": 11216,
+  "tolerance_pct": 50,
+  "value": 2804
+ },
+ "ram.s1.mixwrite.p99us": {
+  "dir": "lower",
+  "floor": 16368,
+  "tolerance_pct": 50,
+  "value": 4092
+ },
+ "ram.s1.msgrate.msgs_sec": {
+  "dir": "higher",
+  "floor": 3356155,
+  "tolerance_pct": 15,
+  "value": 13424620
+ },
+ "ram.s1.query.ops_sec": {
+  "dir": "higher",
+  "floor": 378,
+  "tolerance_pct": 35,
+  "value": 1512
+ },
+ "ram.s1.query.p50us": {
+  "dir": "lower",
+  "floor": 2536,
+  "tolerance_pct": 35,
+  "value": 634
+ },
+ "ram.s1.query.p99us": {
+  "dir": "lower",
+  "floor": 3196,
+  "tolerance_pct": 35,
+  "value": 799
+ },
+ "ram.s1.read.ops_sec": {
+  "dir": "higher",
+  "floor": 407,
+  "tolerance_pct": 35,
+  "value": 1630
+ },
+ "ram.s1.read.p50us": {
+  "dir": "lower",
+  "floor": 2396,
+  "tolerance_pct": 35,
+  "value": 599
+ },
+ "ram.s1.read.p99us": {
+  "dir": "lower",
+  "floor": 3124,
+  "tolerance_pct": 35,
+  "value": 781
+ },
+ "ram.s1.seed.ops_sec": {
+  "dir": "higher",
+  "floor": 64354,
+  "tolerance_pct": 15,
+  "value": 257416
+ },
+ "ram.s1.seed.p50us": {
+  "dir": "lower",
+  "floor": 16,
+  "tolerance_pct": 15,
+  "value": 4
+ },
+ "ram.s1.seed.p99us": {
+  "dir": "lower",
+  "floor": 32,
+  "tolerance_pct": 15,
+  "value": 8
+ },
+ "ram.s1.write.ops_sec": {
+  "dir": "higher",
+  "floor": 718,
+  "tolerance_pct": 15,
+  "value": 2872
+ },
+ "ram.s1.write.p50us": {
+  "dir": "lower",
+  "floor": 2320,
+  "tolerance_pct": 15,
+  "value": 580
+ },
+ "ram.s1.write.p99us": {
+  "dir": "lower",
+  "floor": 3444,
+  "tolerance_pct": 15,
+  "value": 861
+ },
+ "ram.sN.mixread.ops_sec": {
+  "dir": "higher",
+  "floor": 5,
+  "tolerance_pct": 50,
+  "value": 21
+ },
+ "ram.sN.mixread.p50us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "ram.sN.mixread.p99us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "ram.sN.mixwrite.ops_sec": {
+  "dir": "higher",
+  "floor": 0,
+  "tolerance_pct": 50,
+  "value": 2
+ },
+ "ram.sN.mixwrite.p50us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "ram.sN.mixwrite.p99us": {
+  "dir": "lower",
+  "floor": 80000,
+  "tolerance_pct": 50,
+  "value": 20000
+ },
+ "ram.sN.msgrate.msgs_sec": {
+  "dir": "higher",
+  "floor": 611486,
+  "tolerance_pct": 15,
+  "value": 2445944
+ },
+ "ram.sN.query.ops_sec": {
+  "dir": "higher",
+  "floor": 333,
+  "tolerance_pct": 35,
+  "value": 1333
+ },
+ "ram.sN.query.p50us": {
+  "dir": "lower",
+  "floor": 2968,
+  "tolerance_pct": 35,
+  "value": 742
+ },
+ "ram.sN.query.p99us": {
+  "dir": "lower",
+  "floor": 3392,
+  "tolerance_pct": 35,
+  "value": 848
+ },
+ "ram.sN.read.ops_sec": {
+  "dir": "higher",
+  "floor": 372,
+  "tolerance_pct": 35,
+  "value": 1488
+ },
+ "ram.sN.read.p50us": {
+  "dir": "lower",
+  "floor": 2504,
+  "tolerance_pct": 35,
+  "value": 626
+ },
+ "ram.sN.read.p99us": {
+  "dir": "lower",
+  "floor": 3228,
+  "tolerance_pct": 35,
+  "value": 807
+ },
+ "ram.sN.seed.ops_sec": {
+  "dir": "higher",
+  "floor": 74404,
+  "tolerance_pct": 15,
+  "value": 297619
+ },
+ "ram.sN.seed.p50us": {
+  "dir": "lower",
+  "floor": 12,
+  "tolerance_pct": 15,
+  "value": 3
+ },
+ "ram.sN.seed.p99us": {
+  "dir": "lower",
+  "floor": 28,
+  "tolerance_pct": 15,
+  "value": 7
+ },
+ "ram.sN.write.ops_sec": {
+  "dir": "higher",
+  "floor": 634,
+  "tolerance_pct": 15,
+  "value": 2537
+ },
+ "ram.sN.write.p50us": {
+  "dir": "lower",
+  "floor": 2408,
+  "tolerance_pct": 15,
+  "value": 602
+ },
+ "ram.sN.write.p99us": {
+  "dir": "lower",
+  "floor": 3552,
+  "tolerance_pct": 15,
+  "value": 888
+ }
+}
\ No newline at end of file
diff --git a/docs/examples/db-bench/README.md b/docs/examples/db-bench/README.md
index 3b3dd5c..316c693 100644
--- a/docs/examples/db-bench/README.md
+++ b/docs/examples/db-bench/README.md
@@ -40,3 +40,24 @@ compiler classifies the elements OWNED while table refs are scalar ids.
 Query-built multis are runtime-typed and safe. Worked around here
 (single ref local, bucket-major seeding); the compiler fix is its own
 slice.
+
+## Reference-machine numbers (first campaign, 2026-08-21)
+
+`bench/baseline.json` is the contract; headline readings:
+
+- ram seed 257–298k inserts/s; **durable seed ≈4.5k/s** (fsync-per-commit
+  ≈220µs each — the gap iteration 23 exists to close).
+- reads ≈1.5k/s at p50 ≈600µs on a 20k-row store: point lookups are
+  O(table) — the probe walks every slab; index selection never reaches
+  the lookup path. THE read-path finding.
+- mixread 1,280 ops/s single-shard vs **21 ops/s** multi-shard: RPC
+  round-trip × O(table) probes × owner serialization — the arc's honest
+  price until reads index properly.
+- msgrate 13.4M msgs/s same-heap vs 2.45M cross-shard (the mutex-inbox
+  number, stage-2 deviation 4).
+
+## The gate must bite (proven 2026-08-21)
+
+`scripts/db-bench.py --check <results.json>` evaluates a recorded run:
+the real results pass 74/0; a doctored copy (one ops/sec halved) FAILS
+on exactly that metric. Re-run the smoke after any gate change.
diff --git a/docs/stories/language-runtime-database/done/08-shard-actor-runtime.md b/docs/stories/language-runtime-database/done/08-shard-actor-runtime.md
index 438edeb..7334e55 100644
--- a/docs/stories/language-runtime-database/done/08-shard-actor-runtime.md
+++ b/docs/stories/language-runtime-database/done/08-shard-actor-runtime.md
@@ -141,6 +141,19 @@ chain: 1
 
 ## Info
 
+**The arc's measured delta** (iteration 22's first campaign,
+2026-08-21, `bench/baseline.json` — N=20000, 4 mix actors, real-disk
+WO_DATA; single-shard column = the local path, multi-shard = the
+stage-3 RPC):
+
+| metric | WO_SHARDS=1 | default cores | reading |
+| --- | --- | --- | --- |
+| seed inserts/s (ram) | 257,416 | 297,619 | local either way (primary seeds); parity ✓ |
+| seed inserts/s (durable) | 4,492 | 4,478 | fsync-per-commit ≈220µs dominates — the 57× ram gap is iteration 23's case |
+| read ops/s (ram, primary) | 1,630 | 1,488 | point lookups are O(table): the probe walks every slab — the read-path finding |
+| mixread ops/s (actors) | 1,280 | **21** | the RPC price × O(table) probes × owner serialization — the arc's honest cost until reads index properly |
+| msgrate msgs/s | 13,424,620 | 2,445,944 | same-heap vs mutex-inbox: 5.5× — deviation 4's number; rings stay unearned until this is the bottleneck |
+
 **Guarantee contract** (stage-3 refinement 2026-08-21; moved here from
 the slice's marker doc when it landed):
 
diff --git a/docs/superpowers/plans/2026-08-21-db-bench.md b/docs/superpowers/plans/2026-08-21-db-bench.md
index dd4c050..e55443b 100644
--- a/docs/superpowers/plans/2026-08-21-db-bench.md
+++ b/docs/superpowers/plans/2026-08-21-db-bench.md
@@ -183,15 +183,22 @@ there). Story:
   (the arc's recorded delta: single- vs multi-shard columns),
   `docs/examples/db-bench/README.md` (the reference-machine numbers).
 
-- [ ] Full campaign on this machine; inspect the tail; commit the
-  baseline with a message that says it IS the first contract.
-- [ ] Gate-bites smoke (spec acceptance): doctor a copy of the results
-  (halve one ops/sec) and run the gate against it — must FAIL; then the
-  real results — must PASS. Record the procedure in the README.
-- [ ] Copy the arc delta + msgrate numbers into story 8's record and
-  the mutex-inbox note (arc plan deviation 4 references it).
-- [ ] Verify: `just db-bench` exit 0 twice in a row (repeatability);
-  battery. Commit.
+- [x] Full campaign run twice (20 checks/0 fail each incl. restart
+  proofs and 3x kill -9 batteries per shard count); baseline committed
+  as the first contract. DEVIATION: the two-run repeatability check
+  found ~25% jitter on read/query latencies and scheduling-dependent
+  spread on mix* — per-metric tolerances tuned (mix* 50%, read/query
+  35%, rest 15%, rationale in the baseline's _config note); both runs
+  pass the tuned contract 74/0.
+- [x] Gate-bites proven: `--check` on a doctored copy (one ops/sec
+  halved) FAILS on exactly that metric; both real runs PASS. Procedure
+  in the README (a `--check` gate-only mode was added for this).
+- [x] Arc delta + msgrate recorded: story 8 carries the measured table
+  (durable seed 4.5k/s vs ram 297k/s = 23's case; mixread 1280 vs 21
+  ops/s = the RPC x O(table)-probe price; msgrate 13.4M vs 2.45M =
+  deviation 4's mutex-inbox number). Headline findings in the sample
+  README: point lookups are O(table) — the read-path finding.
+- [x] Verified: two full campaigns green; battery green. Commit.
 
 ## Task 6 — closeout
 
diff --git a/scripts/db-bench.py b/scripts/db-bench.py
index aa323a7..8438b6c 100755
--- a/scripts/db-bench.py
+++ b/scripts/db-bench.py
@@ -219,6 +219,14 @@ def write_baseline(metrics):
     ok(f"baseline written ({len(base) - 1} metrics)")
 
 def main():
+    # --check <results.json>: gate-only evaluation of a recorded run — the
+    # gate-bites smoke doctors a copy and this mode must FAIL on it
+    if "--check" in sys.argv:
+        f = sys.argv[sys.argv.index("--check") + 1]
+        gate(json.load(open(f)))
+        print()
+        print(f"db-bench --check: {len(passed) + len(failed)} checks, {len(failed)} failures")
+        sys.exit(1 if failed else 0)
     build()
     metrics = campaign()
     durability(metrics)