6cb0140fae
The public history carried kat/__pycache__/mlkem768_reference.cpython-314.pyc, a compiled Python artifact embedding the operator's absolute local path. Text secret scanners do not read compiled binaries, which is exactly how it slipped through, and removing it from the tip would have left it reachable through the old root commits. So this repository is republished from a single clean root. This root also carries, from the previously unpublished line of work: - corrected LICENSE year, LICENSING.md, VERIFY-POLICY.md, and CITATIONS-UNRESOLVED.md remeasured 2026-08-11 (101 paths, README aligned) - O-018: run_consensus_verification.py ran 19 of 29 models and reported PASS; it now runs all 29, and computemarket_smt.py gains resolveByTimeout / reclaimUnsettled cases plus a negative control - O-006: the word 'audited' removed from next to Bouncy Castle, twice, after a concurrent edit resurrected it - O-014: prior art named and dated - Algorand's native falcon_verify shipped about ten months before AERE's precompiles; the primacy claim is withdrawn where it was implied - bench/ scripts parametrized so they actually run for an outsider (the earlier textual sanitization left $STAGING unexpanded inside Python strings) - AIP-2/AIP-3 errata with measured figures, spec remeasurements at 2026-08-01, and the spec-zk-stack retractions (owner is an operational key, not the Foundation; 'maximally sound' withdrawn; aggregator V1 deprecated) The redacted bench-host environment files from the sanitized line are kept exactly as published; the unredacted local variants are not carried.
52 lines
2.6 KiB
Plaintext
52 lines
2.6 KiB
Plaintext
=======================================================
|
|
BLOCK-STM SPEEDUP BENCHMARK
|
|
machine threads available: 16
|
|
txns per batch : 20000
|
|
simulated EVM gas per tx : 80 keccak-rounds
|
|
timing : median of 7 runs (ms)
|
|
=======================================================
|
|
|
|
### LOW-CONFLICT (50k accounts, few hot, ~2% hot-touch)
|
|
parallel==sequential check: PASS
|
|
sequential baseline : 663.24 ms
|
|
threads time(ms) speedup vs-1thread aborts abort%
|
|
1 725.04 0.91x 1.00x 0 0.0%
|
|
2 380.42 1.74x 1.91x 3 0.0%
|
|
4 204.86 3.24x 3.54x 13 0.1%
|
|
8 116.83 5.68x 6.21x 36 0.2%
|
|
16 79.90 8.30x 9.07x 61 0.3%
|
|
|
|
### MEDIUM-CONFLICT (5k accounts, ~15% hot-touch)
|
|
parallel==sequential check: PASS
|
|
sequential baseline : 661.01 ms
|
|
threads time(ms) speedup vs-1thread aborts abort%
|
|
1 705.12 0.94x 1.00x 0 0.0%
|
|
2 363.28 1.82x 1.94x 37 0.2%
|
|
4 189.85 3.48x 3.71x 90 0.4%
|
|
8 107.00 6.18x 6.59x 198 1.0%
|
|
16 68.74 9.62x 10.26x 423 2.1%
|
|
|
|
### HIGH-CONFLICT (single hot counter, ~80% hot-touch)
|
|
parallel==sequential check: PASS
|
|
sequential baseline : 669.68 ms
|
|
threads time(ms) speedup vs-1thread aborts abort%
|
|
1 694.53 0.96x 1.00x 0 0.0%
|
|
2 385.06 1.74x 1.80x 1335 6.2%
|
|
4 216.35 3.10x 3.21x 3425 13.5%
|
|
8 134.93 4.96x 5.15x 6940 19.8%
|
|
16 130.67 5.13x 5.32x 13314 23.8%
|
|
|
|
### PATHOLOGICAL (every tx increments the SAME slot: no parallelism)
|
|
parallel==sequential check: PASS
|
|
sequential baseline : 676.98 ms
|
|
threads time(ms) speedup vs-1thread aborts abort%
|
|
1 694.65 0.97x 1.00x 0 0.0%
|
|
2 727.81 0.93x 0.95x 13410 1.6%
|
|
4 783.44 0.86x 0.89x 22510 5.8%
|
|
8 824.31 0.82x 0.84x 27712 12.4%
|
|
16 884.96 0.76x 0.78x 30654 18.4%
|
|
|
|
Note: speedup is vs the sequential oracle. High-conflict batches
|
|
serialize on the hot slot and are expected to show little or no
|
|
speedup -- that is the honest, correct behavior of Block-STM.
|