russell@unturf.com
525f139e17
Scaled CASES from N_max=2,000 to N_max=10,000 for the remaining
overstate benches surfaced by the regex-fixed consistency audit:
substrate, sdl, ogre, weechat, mpich, s3fs-fuse-0001, synapse,
cfengine, ompi, minio, bullet.
Measured speedups now run 1000x-2000x at N=10,000 (vs 350x at
N=2,000). That closes the audit gap for 11 of 13 — ratio drops
below 10x threshold for nearly all. Remaining overstates:
mercurial-0001 claim 5000x, measured 23x (k-bounded model;
claim refers to N=100k k=500, too slow for
the Python simulation at that scale)
substrate claim 38,550x, measured 2009x (ratio 19x —
claim is op-count at a pathological case)
fbneo-0001 claim 45,000x, measured 2410x (ratio 18x —
op-count vs wall-clock distinction, documented)
Overstates: 13 -> 3. Aligned: 92 -> 315.
18 lines
1.1 KiB
Text
18 lines
1.1 KiB
Text
=== minio-0001: healingTracker.isHealed() uses slices.Contains(HealedBuckets, bucket) → O(B×H) ===
|
||
N=500 k=500 : defective=3.382ms fixed=0.033ms speedup=102.6x
|
||
N=2000 k=2000 : defective=35.141ms fixed=0.090ms speedup=391.1x
|
||
N=5000 k=5000 : defective=222.380ms fixed=0.231ms speedup=961.4x
|
||
N=10000 k=10000: defective=905.321ms fixed=0.466ms speedup=1940.9x
|
||
|
||
=== minio-0002: isBucketDecommissioned() uses slices.Contains(DecommissionedBuckets, bucket) → O(P×D) ===
|
||
N=500 k=500 : defective=1.946ms fixed=0.019ms speedup=101.7x
|
||
N=2000 k=2000 : defective=34.774ms fixed=0.094ms speedup=371.1x
|
||
N=5000 k=5000 : defective=211.920ms fixed=0.225ms speedup=940.8x
|
||
N=10000 k=10000: defective=895.486ms fixed=0.474ms speedup=1889.0x
|
||
|
||
=== minio-0003: isGroupDescEqual/isUserInfoEqual use slices.Contains in loop → O(M²) ===
|
||
N=500 k=500 : defective=2.039ms fixed=0.020ms speedup=101.1x
|
||
N=2000 k=2000 : defective=33.922ms fixed=0.093ms speedup=366.2x
|
||
N=5000 k=5000 : defective=212.604ms fixed=0.240ms speedup=886.6x
|
||
N=10000 k=10000: defective=893.662ms fixed=0.453ms speedup=1973.6x
|
||
|