diff --git a/docs/architecture/cells.md b/docs/architecture/cells.md index ff46557fb8..d57040ff98 100644 --- a/docs/architecture/cells.md +++ b/docs/architecture/cells.md @@ -967,11 +967,10 @@ read-only scope, lists any tenancy row the directory does not place. `enforced` with the orgs held. Every Loom write on a cell is marked, so it pays the setting lookup alone, measured on the statements with the triggers on and off (no cost distinguishable from zero: six rounds range from −0.36 to 0.49 µs a row, [benchmark - results](../benchmarks/results.md)), which predicts the gate holds by a wide margin. The - gate itself is an end-to-end `memory.set` and delivery-pass measurement on a cell, which - needs attach and is checked with it; until then the triggers, which a later cell gets from - prepare's next migrating boot, ship on the statement-level measurement. `loom.tenancy` - itself has its own guard, since a tenancy row makes an org held: `tenancy_record`, a + results](../benchmarks/results.md)). These statement timings do not establish either + gate. Both require the end-to-end `memory.set` and delivery-pass acceptance workload on + an attached cell. `loom.tenancy` itself has its own guard, since a tenancy row makes an + org held: `tenancy_record`, a row-level `AFTER INSERT OR UPDATE OF org` origin trigger with the same marker `WHEN`, calls `loom.tenancy_refuse_unmarked()` (migration 012), which refuses, `LT001`, an unmarked session's tenancy row, or an update that moves one to another org, in every state diff --git a/docs/benchmarks/results.md b/docs/benchmarks/results.md index fd940bf8ab..7d7e65c247 100644 --- a/docs/benchmarks/results.md +++ b/docs/benchmarks/results.md @@ -1216,11 +1216,10 @@ host dominate the single-row figures, so they show no measurable cost, not a gai and the cost is the difference within the round. A marked session, which every Loom process with `LOOM_CELL` set opens, shows no cost the rounds can tell from zero (the six rounds' differences range from −0.36 to 0.49 µs a row); an unmarked one pays 12 to 14 µs -a row for the state read and the `loom.tenancy` lookup. Loom's own writes are marked, so -this predicts the ship gate (a `memory.set` p99 rise of at most 10 ms, a delivery pass -at most 10% slower) holds by a wide margin; the cost lands on scripts and older builds -writing a cell without the check. This measures the triggers on the statements, not a -whole `memory.set` through a server on a cell; the end-to-end gate follows. +a row for the state read and the `loom.tenancy` lookup. These statement timings do not +establish the `memory.set` p99 or delivery-throughput gates. Those require the end-to-end +acceptance workload below. An unmarked session, including a script or older build writing +a cell without the check, pays the measured state-read and tenancy-lookup cost. A small local diagnostic on an attached first cell records 400 sequential `memory.set` calls per run and 1,000-row statement inserts from marked and unmarked sessions, three