{ "Schema": "syrf.project-statistics.baseline/v1", "Measurement": "ScreeningSourceWrites", "DatasetId": "PS-WRITE-01", "DatasetName": "Bounded screening contention corpus", "Seed": 0, "Scale": 1, "GeneratorVersion": "screening-write-v1", "ContainerImage": "mongo:8.0", "ServerVersion": "8.0.28", "HostProcessorCount": 48, "RuntimeVersion": "10.0.0", "OperatingSystem": "Unix 6.8.0.124", "ContainerResourceCapsApplied": false, "WarmupIterations": 10, "MeasuredIterations": 100, "StartedAtUtc": "2026-09-30T07:03:59.1385659Z", "CompletedAtUtc": "2026-09-30T07:05:55.9860124Z", "Corpus": { "Projects": 1, "Studies": 10, "StudiesInTargetProject": 10, "ScreeningDecisions": 10, "AnnotationSessions": 0, "Annotations": 0, "SystematicSearches": 0, "Reviewers": 10, "Stages": 1, "AnnotationQuestions": 0, "DistinctTallyKeysInTargetProject": 0 }, "Durations": [ { "Metric": "same_study_1_reviewers_source_only", "WarmupIterations": 10, "Iterations": 100, "MinMs": 3.75, "P50Ms": 7.2678, "P95Ms": 8.6244, "MaxMs": 25.2326, "MeanMs": 7.368874 }, { "Metric": "same_study_1_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 100, "MinMs": 2.505, "P50Ms": 5.064, "P95Ms": 6.1586, "MaxMs": 22.6546, "MeanMs": 5.1781289999999975 }, { "Metric": "same_study_1_reviewers_materialized", "WarmupIterations": 10, "Iterations": 100, "MinMs": 22.3035, "P50Ms": 37.7891, "P95Ms": 61.3579, "MaxMs": 67.6369, "MeanMs": 39.47670600000001 }, { "Metric": "same_study_1_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 100, "MinMs": 18.5828, "P50Ms": 31.9072, "P95Ms": 51.4478, "MaxMs": 57.9328, "MeanMs": 33.22837 }, { "Metric": "same_study_2_reviewers_materialized", "WarmupIterations": 10, "Iterations": 200, "MinMs": 5.8638, "P50Ms": 31.2407, "P95Ms": 53.7368, "MaxMs": 63.828, "MeanMs": 31.866394000000017 }, { "Metric": "same_study_2_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 374, "MinMs": 2.8861, "P50Ms": 6.8249, "P95Ms": 38.9985, "MaxMs": 47.3954, "MeanMs": 13.387299732620317 }, { "Metric": "same_study_2_reviewers_source_only", "WarmupIterations": 10, "Iterations": 200, "MinMs": 3.8427, "P50Ms": 8.6216, "P95Ms": 13.6623, "MaxMs": 20.5159, "MeanMs": 8.644556000000001 }, { "Metric": "same_study_2_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 300, "MinMs": 2.3788, "P50Ms": 4.1654, "P95Ms": 5.8782, "MaxMs": 13.7351, "MeanMs": 4.375798666666668 }, { "Metric": "same_study_5_reviewers_source_only", "WarmupIterations": 10, "Iterations": 500, "MinMs": 3.4083, "P50Ms": 16.5962, "P95Ms": 22.8228, "MaxMs": 31.5367, "MeanMs": 15.074673200000005 }, { "Metric": "same_study_5_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 1200, "MinMs": 2.3315, "P50Ms": 4.6976, "P95Ms": 6.2152, "MaxMs": 15.6343, "MeanMs": 4.888834749999998 }, { "Metric": "same_study_5_reviewers_materialized", "WarmupIterations": 10, "Iterations": 500, "MinMs": 6.5185, "P50Ms": 29.4699, "P95Ms": 51.9652, "MaxMs": 71.8013, "MeanMs": 31.0219048 }, { "Metric": "same_study_5_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 1267, "MinMs": 2.3098, "P50Ms": 6.6454, "P95Ms": 33.3264, "MaxMs": 51.5849, "MeanMs": 8.881071428571428 }, { "Metric": "same_study_10_reviewers_materialized", "WarmupIterations": 10, "Iterations": 1000, "MinMs": 6.7009, "P50Ms": 31.9373, "P95Ms": 48.5405, "MaxMs": 89.936, "MeanMs": 33.381029799999986 }, { "Metric": "same_study_10_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 2761, "MinMs": 2.3112, "P50Ms": 6.8753, "P95Ms": 18.3263, "MaxMs": 55.7372, "MeanMs": 8.202834878667144 }, { "Metric": "same_study_10_reviewers_source_only", "WarmupIterations": 10, "Iterations": 1000, "MinMs": 4.1106, "P50Ms": 20.0885, "P95Ms": 32.0429, "MaxMs": 36.8821, "MeanMs": 19.76120820000002 }, { "Metric": "same_study_10_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 2694, "MinMs": 2.2913, "P50Ms": 5.2944, "P95Ms": 7.1037, "MaxMs": 20.6789, "MeanMs": 5.612548181143268 }, { "Metric": "different_studies_1_reviewers_source_only", "WarmupIterations": 10, "Iterations": 100, "MinMs": 4.0031, "P50Ms": 5.5794, "P95Ms": 6.5305, "MaxMs": 16.0961, "MeanMs": 5.730740000000001 }, { "Metric": "different_studies_1_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 100, "MinMs": 2.8402, "P50Ms": 4.0844, "P95Ms": 4.9422, "MaxMs": 12.7966, "MeanMs": 4.136688999999999 }, { "Metric": "different_studies_1_reviewers_materialized", "WarmupIterations": 10, "Iterations": 100, "MinMs": 17.5011, "P50Ms": 41.3744, "P95Ms": 49.7942, "MaxMs": 58.1607, "MeanMs": 38.889327 }, { "Metric": "different_studies_1_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 100, "MinMs": 14.8928, "P50Ms": 35.6463, "P95Ms": 41.9919, "MaxMs": 47.1189, "MeanMs": 32.924049000000004 }, { "Metric": "different_studies_2_reviewers_materialized", "WarmupIterations": 10, "Iterations": 200, "MinMs": 17.765, "P50Ms": 47.2517, "P95Ms": 101.5322, "MaxMs": 110.3024, "MeanMs": 62.182419999999986 }, { "Metric": "different_studies_2_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 400, "MinMs": 9.3381, "P50Ms": 25.3146, "P95Ms": 40.5858, "MaxMs": 48.865, "MeanMs": 27.374023499999993 }, { "Metric": "different_studies_2_reviewers_source_only", "WarmupIterations": 10, "Iterations": 200, "MinMs": 3.7971, "P50Ms": 5.3, "P95Ms": 6.6533, "MaxMs": 15.5199, "MeanMs": 5.463906999999998 }, { "Metric": "different_studies_2_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 200, "MinMs": 2.768, "P50Ms": 3.9338, "P95Ms": 5.1176, "MaxMs": 13.9971, "MeanMs": 4.1432395000000035 }, { "Metric": "different_studies_5_reviewers_source_only", "WarmupIterations": 10, "Iterations": 500, "MinMs": 3.7155, "P50Ms": 5.6258, "P95Ms": 6.6887, "MaxMs": 17.6103, "MeanMs": 5.9372035999999975 }, { "Metric": "different_studies_5_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 500, "MinMs": 2.3724, "P50Ms": 4.3851, "P95Ms": 5.1612, "MaxMs": 16.3755, "MeanMs": 4.6638216 }, { "Metric": "different_studies_5_reviewers_materialized", "WarmupIterations": 10, "Iterations": 500, "MinMs": 6.6064, "P50Ms": 44.3267, "P95Ms": 98.0474, "MaxMs": 111.8515, "MeanMs": 48.763931 }, { "Metric": "different_studies_5_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 998, "MinMs": 3.3888, "P50Ms": 20.3537, "P95Ms": 39.013, "MaxMs": 56.44, "MeanMs": 20.658513026052116 }, { "Metric": "different_studies_10_reviewers_materialized", "WarmupIterations": 10, "Iterations": 1000, "MinMs": 6.9361, "P50Ms": 51.2885, "P95Ms": 97.0198, "MaxMs": 120.4311, "MeanMs": 51.79774079999995 }, { "Metric": "different_studies_10_reviewers_materialized_attempt", "WarmupIterations": 10, "Iterations": 1999, "MinMs": 3.1896, "P50Ms": 21.6239, "P95Ms": 38.214, "MaxMs": 53.6178, "MeanMs": 21.208481590795362 }, { "Metric": "different_studies_10_reviewers_source_only", "WarmupIterations": 10, "Iterations": 1000, "MinMs": 2.4495, "P50Ms": 5.0072, "P95Ms": 8.1643, "MaxMs": 29.3423, "MeanMs": 5.6105395 }, { "Metric": "different_studies_10_reviewers_source_only_attempt", "WarmupIterations": 10, "Iterations": 1000, "MinMs": 1.7774, "P50Ms": 3.8702, "P95Ms": 5.2858, "MaxMs": 28.0998, "MeanMs": 4.209143200000001 } ], "RoundTrips": [], "Notes": { "scope": "Capacity-save source integration, active-reviewer tracking off, screening family only; no HTTP, assignment, SignalR or reviewer-maintenance claim.", "controller_boundary": "Current ReviewController resolves durable tracking mode per attempt and selects the plain save when tracking is off. This harness retains the capacity-save repository overload with tracking off, so its p95 is not the current HTTP path or reviewer-maintenance acceptance.", "sampling": "Per submission includes source reload, envelope/receipt read, classification and up to three fresh-source save attempts. Attempt timings cover repository save only. Warmup excluded. Same reviewers edit existing decisions after first round.", "unknown_commit": "Driver CommandFailed events count UnknownTransactionCommitResult labels; CommitTransaction command starts count actual commit attempts. No fault injection. Other exceptions fail the run.", "timing_gate": "Report candidate/baseline p95 overhead against \u003C10%; timing never determines CI correctness on a shared host. Exhaustion remains an observed outcome, not a successful save.", "round_trip_counting": "Command counting is opt-in via SYRF_STATS_COUNT_ROUNDTRIPS; it is OFF for this run. A CommandStartedEvent subscriber forces the driver to materialize every command document, and the materialized arm issues far more commands than the source-only arm, so the subscriber\u0027s own cost is asymmetric. Run once with it off to get a no-subscriber control timing, and once with it on for the command breakdown. With it off, commands and commit_attempts are recorded as not_counted and no per-command-name record is emitted.", "per_attempt_commands": "Per-attempt command counts are recorded only for the single-reviewer cells. With two or more reviewers the submissions run concurrently against one shared counter, so a per-attempt delta would include another reviewer\u0027s commands and is deliberately not claimed.", "same_study_1_reviewers_source_only_outcomes": "saved=100; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=100; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "same_study_1_reviewers_materialized_outcomes": "saved=100; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=0; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1199", "same_study_1_reviewers_p95_gate": "overhead_percent=611.45; passes_less_than_10_percent=False", "same_study_2_reviewers_materialized_outcomes": "saved=126; version_conflicts=248; exhausted=74; parity_batches=100; no_delta_saves=26; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1273", "same_study_2_reviewers_source_only_outcomes": "saved=200; version_conflicts=100; exhausted=0; parity_batches=100; no_delta_saves=200; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "same_study_2_reviewers_p95_gate": "overhead_percent=293.32; passes_less_than_10_percent=False", "same_study_5_reviewers_source_only_outcomes": "saved=300; version_conflicts=900; exhausted=200; parity_batches=100; no_delta_saves=300; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "same_study_5_reviewers_materialized_outcomes": "saved=147; version_conflicts=1120; exhausted=353; parity_batches=100; no_delta_saves=48; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1805", "same_study_5_reviewers_p95_gate": "overhead_percent=127.69; passes_less_than_10_percent=False", "same_study_10_reviewers_materialized_outcomes": "saved=157; version_conflicts=2604; exhausted=843; parity_batches=100; no_delta_saves=62; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=2111", "same_study_10_reviewers_source_only_outcomes": "saved=303; version_conflicts=2391; exhausted=697; parity_batches=100; no_delta_saves=303; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "same_study_10_reviewers_p95_gate": "overhead_percent=51.49; passes_less_than_10_percent=False", "different_studies_1_reviewers_source_only_outcomes": "saved=100; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=100; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "different_studies_1_reviewers_materialized_outcomes": "saved=100; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=0; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1199", "different_studies_1_reviewers_p95_gate": "overhead_percent=662.49; passes_less_than_10_percent=False", "different_studies_2_reviewers_materialized_outcomes": "saved=200; version_conflicts=200; exhausted=0; parity_batches=100; no_delta_saves=0; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1199", "different_studies_2_reviewers_source_only_outcomes": "saved=200; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=200; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "different_studies_2_reviewers_p95_gate": "overhead_percent=1426.04; passes_less_than_10_percent=False", "different_studies_5_reviewers_source_only_outcomes": "saved=500; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=500; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "different_studies_5_reviewers_materialized_outcomes": "saved=350; version_conflicts=648; exhausted=150; parity_batches=100; no_delta_saves=150; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1199", "different_studies_5_reviewers_p95_gate": "overhead_percent=1365.87; passes_less_than_10_percent=False", "different_studies_10_reviewers_materialized_outcomes": "saved=601; version_conflicts=1398; exhausted=399; parity_batches=100; no_delta_saves=398; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=1199", "different_studies_10_reviewers_source_only_outcomes": "saved=1000; version_conflicts=0; exhausted=0; parity_batches=100; no_delta_saves=1000; commands=not_counted; commit_attempts=not_counted; unknown_commit_failures=0; max_summary_bytes=0", "different_studies_10_reviewers_p95_gate": "overhead_percent=1088.34; passes_less_than_10_percent=False" } }