Log in
benchmarksPublic

Commit 4310ddbe

· Browse files at this commit

4310ddbe60b6

Merged on mrmrs-10a6b5
mrmrs · 1 month ago · 3 files ~3 modified
M README.md
@@ -346,7 +346,7 @@
346346 --materialization-baseline-subject oak_baseline \
347347 --scenarios legacy_selected_clone_materialization_v1 \
348348 --runs 10 --randomize-subject-order --random-seed 46802 \
349- --claim-materialization-timing \
349+ --materialization-timing-controls-met \
350350 --workdir /abs/worker-owned/materialization-ab/work \
351351 --results /abs/worker-owned/materialization-ab/results
352352 ```
@@ -358,10 +358,12 @@
358358 The scenario records raw proxy byte/connection totals, but its correctness gate
359359 uses normalized pull phases, response/object/chunk counters, and connection
360360 count; it does not pretend variable HTTP framing is product work. Clone, first
361-post-clone status, and first no-op commit are separate rows. Timing is marked
362-claimable only when `--claim-materialization-timing` is explicit, the shared
363-lock is held, order is randomized, and at least ten runs are present; operators
364-must still coordinate the documented quiet-host window.
361+post-clone status, and first no-op commit are separate rows. The harness marks
362+only that its local timing controls are met when
363+`--materialization-timing-controls-met` is explicit, the shared lock is held,
364+order is randomized, and at least ten runs are present. That marker is not a
365+standalone speedup claim: the documented quiet-host window and a separate
366+interleaved A/A noise campaign are still required evidence.
365367
366368 These cells are different evidence classes. Local Git wall time is not a
367369 network comparison, and the Oak local-server row is loopback-only. Its second