EGraph hash map optimizations

[jonathanvdc]

No description provided.

[github-actions]

JMH comparison (baseline vs PR)

Benchmark	Params	Baseline	PR	Δ (PR/Base)	Unit
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=false, size=500, threadCount=1	195.915	194.02	0.990× (-1.0%)	ms/op
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=false, size=500, threadCount=2	314.877	297.897	0.946× (-5.4%)	ms/op
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=true, size=500, threadCount=1	7.84266	8.22159	1.048× (+4.8%)	ms/op
IncrementalBenchmarks.incrementalPolynomial	depth=6, mutableEGraph=true, size=500, threadCount=2	72.2722	72.8113	1.007× (+0.7%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=false, size=500, threadCount=1	347.304	325.479	0.937× (-6.3%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=false, size=500, threadCount=2	506.175	506.636	1.001× (+0.1%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=true, size=500, threadCount=1	138.269	138.692	1.003× (+0.3%)	ms/op
IncrementalBenchmarks.oneByOnePolynomial	depth=6, mutableEGraph=true, size=500, threadCount=2	303.459	296	0.975× (-2.5%)	ms/op
LiarBenchmarks.findGemmInMm	threadCount=1	230.015	223.56	0.972× (-2.8%)	ms/op
LiarBenchmarks.findGemmInMm	threadCount=2	182.511	173.656	0.951× (-4.9%)	ms/op
LiarBenchmarks.findGemvInMv	threadCount=1	60.7702	61.8665	1.018× (+1.8%)	ms/op
LiarBenchmarks.findGemvInMv	threadCount=2	55.1882	56.5584	1.025× (+2.5%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=20, threadCount=1	4.45602	4.48606	1.007× (+0.7%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=20, threadCount=2	4.28422	4.20038	0.980× (-2.0%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=40, threadCount=1	30.1247	29.9196	0.993× (-0.7%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=40, threadCount=2	23.5948	23.7014	1.005× (+0.5%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=80, threadCount=1	267.391	261.32	0.977× (-2.3%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=false, size=80, threadCount=2	176.012	157.814	0.897× (-10.3%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=20, threadCount=1	2.721	2.7593	1.014× (+1.4%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=20, threadCount=2	2.55217	2.6941	1.056× (+5.6%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=40, threadCount=1	20.4907	20.678	1.009× (+0.9%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=40, threadCount=2	15.2617	15.5157	1.017× (+1.7%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=80, threadCount=1	176.152	168.795	0.958× (-4.2%)	ms/op
MatmulBenchmarks.nmm	mutableEGraph=true, size=80, threadCount=2	111.13	107.605	0.968× (-3.2%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=5, threadCount=1	87.1344	86.1413	0.989× (-1.1%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=5, threadCount=2	80.3963	79.6796	0.991× (-0.9%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=6, threadCount=1	408.011	403.221	0.988× (-1.2%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=false, size=6, threadCount=2	367.177	356.347	0.971× (-2.9%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=5, threadCount=1	34.2072	34.2745	1.002× (+0.2%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=5, threadCount=2	28.7418	28.2641	0.983× (-1.7%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=6, threadCount=1	139.329	141.9	1.018× (+1.8%)	ms/op
PolyBenchmarks.polynomial	mutableEGraph=true, size=6, threadCount=2	113.261	113.056	0.998× (-0.2%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=false, threadCount=1	34807.5	33992.6	0.977× (-2.3%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=false, threadCount=2	33044.5	32788.2	0.992× (-0.8%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=true, threadCount=1	4544.92	4415.7	0.972× (-2.8%)	ms/op
VectorBenchmarks.blinnPhong	mutableEGraph=true, threadCount=2	3293.88	3111.63	0.945× (-5.5%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=false, threadCount=1	467.809	475.573	1.017× (+1.7%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=false, threadCount=2	397.789	387.364	0.974× (-2.6%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=true, threadCount=1	170.632	179.746	1.053× (+5.3%)	ms/op
VectorBenchmarks.gramSchmidt	mutableEGraph=true, threadCount=2	128.363	131.879	1.027× (+2.7%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=false, threadCount=1	292.259	293.892	1.006× (+0.6%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=false, threadCount=2	247.956	239.184	0.965× (-3.5%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=true, threadCount=1	120.079	115.464	0.962× (-3.8%)	ms/op
VectorBenchmarks.reflection	mutableEGraph=true, threadCount=2	85.8107	83.7479	0.976× (-2.4%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=false, threadCount=1	3.90033	3.73062	0.956× (-4.4%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=false, threadCount=2	4.26656	4.28576	1.004× (+0.4%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=true, threadCount=1	1.6568	1.67698	1.012× (+1.2%)	ms/op
VectorBenchmarks.vectorNormalization	mutableEGraph=true, threadCount=2	2.15059	2.30263	1.071× (+7.1%)	ms/op
Geomean	threadCount=1, mutableEGraph=false	—	—	0.986× (-1.4%)
Geomean	threadCount=1, mutableEGraph=true	—	—	1.004× (+0.4%)
Geomean	threadCount=2, mutableEGraph=false	—	—	0.977× (-2.3%)
Geomean	threadCount=2, mutableEGraph=true	—	—	1.001× (+0.1%)

Note: < 1.0× means faster on PR; > 1.0× means slower.