Skip to content

Latest commit

 

History

History
208 lines (187 loc) · 26.4 KB

bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67-vs-3.10.4.md

File metadata and controls

208 lines (187 loc) · 26.4 KB

Results vs. 3.10.4

  • fork: python
  • ref: 29f8a67ae00081a36fdc
  • machine: linux-x86_64
  • commit hash: 29f8a67
  • commit date: 2025-02-08
  • overall geometric mean: 1.329x faster
  • HPT reliability: 100.00%
  • HPT 99th percentile: 1.21x faster
  • Memory change: 1.29x

Benchmarks with tag 'apps':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
2to3 350 ms 290 ms: 1.21x faster
docutils 3.41 sec 2.95 sec: 1.16x faster
html5lib 94.6 ms 67.9 ms: 1.39x faster
Geometric mean (ref) 1.25x faster

Benchmarks with tag 'asyncio':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
async_tree_io 1.60 sec 650 ms: 2.46x faster
async_tree_none 692 ms 289 ms: 2.39x faster
async_tree_memoization 819 ms 352 ms: 2.33x faster
async_tree_cpu_io_mixed 936 ms 522 ms: 1.79x faster
Geometric mean (ref) 2.23x faster

Benchmarks with tag 'math':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
float 111 ms 70.1 ms: 1.58x faster
nbody 134 ms 101 ms: 1.33x faster
pidigits 271 ms 254 ms: 1.07x faster
Geometric mean (ref) 1.31x faster

Benchmarks with tag 'regex':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
regex_compile 190 ms 137 ms: 1.38x faster
regex_v8 27.2 ms 25.0 ms: 1.09x faster
regex_dna 261 ms 241 ms: 1.08x faster
regex_effbot 3.09 ms 3.11 ms: 1.01x slower
Geometric mean (ref) 1.13x faster

Benchmarks with tag 'serialize':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
unpickle_pure_python 312 us 206 us: 1.52x faster
tomli_loads 2.92 sec 2.07 sec: 1.41x faster
xml_etree_process 75.9 ms 55.5 ms: 1.37x faster
pickle_pure_python 455 us 334 us: 1.36x faster
json_dumps 14.5 ms 11.8 ms: 1.24x faster
xml_etree_parse 160 ms 136 ms: 1.18x faster
xml_etree_generate 92.3 ms 79.3 ms: 1.16x faster
xml_etree_iterparse 110 ms 95.9 ms: 1.15x faster
json_loads 30.3 us 26.6 us: 1.14x faster
unpickle_list 4.65 us 4.83 us: 1.04x slower
unpickle 13.5 us 14.2 us: 1.05x slower
pickle 9.89 us 12.1 us: 1.23x slower
pickle_dict 29.5 us 37.3 us: 1.26x slower
pickle_list 4.12 us 5.36 us: 1.30x slower
Geometric mean (ref) 1.10x faster

Benchmarks with tag 'startup':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
python_startup_no_site 7.33 ms 9.01 ms: 1.23x slower
python_startup 11.5 ms 16.0 ms: 1.39x slower
Geometric mean (ref) 1.31x slower

Benchmarks with tag 'template':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
mako 14.7 ms 9.79 ms: 1.50x faster
django_template 50.2 ms 36.1 ms: 1.39x faster
genshi_text 31.4 ms 24.2 ms: 1.30x faster
genshi_xml 63.3 ms 56.6 ms: 1.12x faster
Geometric mean (ref) 1.32x faster

All benchmarks:

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67
typing_runtime_protocols 537 us 180 us: 2.98x faster
async_tree_io 1.60 sec 650 ms: 2.46x faster
async_tree_none 692 ms 289 ms: 2.39x faster
async_tree_memoization 819 ms 352 ms: 2.33x faster
deltablue 7.50 ms 3.45 ms: 2.18x faster
asyncio_tcp 779 ms 364 ms: 2.14x faster
generators 57.3 ms 29.1 ms: 1.97x faster
asyncio_tcp_ssl 3.10 sec 1.58 sec: 1.96x faster
async_tree_cpu_io_mixed 936 ms 522 ms: 1.79x faster
chaos 109 ms 61.0 ms: 1.78x faster
scimark_lu 167 ms 94.6 ms: 1.76x faster
pylint 566 ms 322 ms: 1.76x faster
scimark_monte_carlo 107 ms 61.1 ms: 1.76x faster
richards_super 90.6 ms 52.1 ms: 1.74x faster
go 262 ms 151 ms: 1.73x faster
logging_silent 167 ns 97.4 ns: 1.72x faster
deepcopy_memo 49.8 us 29.4 us: 1.69x faster
raytrace 489 ms 296 ms: 1.65x faster
scimark_sor 180 ms 109 ms: 1.65x faster
spectral_norm 139 ms 84.8 ms: 1.64x faster
richards 75.1 ms 45.8 ms: 1.64x faster
pyflate 733 ms 447 ms: 1.64x faster
deepcopy 468 us 286 us: 1.64x faster
sqlglot_parse 2.24 ms 1.40 ms: 1.59x faster
float 111 ms 70.1 ms: 1.58x faster
crypto_pyaes 119 ms 77.5 ms: 1.54x faster
unpickle_pure_python 312 us 206 us: 1.52x faster
mako 14.7 ms 9.79 ms: 1.50x faster
sqlglot_transpile 2.68 ms 1.79 ms: 1.50x faster
coroutines 30.3 ms 20.8 ms: 1.46x faster
logging_simple 8.88 us 6.23 us: 1.42x faster
logging_format 9.64 us 6.77 us: 1.42x faster
tomli_loads 2.92 sec 2.07 sec: 1.41x faster
html5lib 94.6 ms 67.9 ms: 1.39x faster
comprehensions 26.7 us 19.2 us: 1.39x faster
django_template 50.2 ms 36.1 ms: 1.39x faster
deepcopy_reduce 4.01 us 2.89 us: 1.39x faster
regex_compile 190 ms 137 ms: 1.38x faster
xml_etree_process 75.9 ms 55.5 ms: 1.37x faster
pickle_pure_python 455 us 334 us: 1.36x faster
thrift 1.18 ms 872 us: 1.35x faster
hexiom 9.42 ms 7.05 ms: 1.34x faster
nbody 134 ms 101 ms: 1.33x faster
pycparser 1.67 sec 1.26 sec: 1.33x faster
sqlalchemy_declarative 190 ms 146 ms: 1.30x faster
genshi_text 31.4 ms 24.2 ms: 1.30x faster
pprint_pformat 2.15 sec 1.68 sec: 1.28x faster
pprint_safe_repr 1.05 sec 819 ms: 1.28x faster
pathlib 21.4 ms 16.7 ms: 1.28x faster
sqlalchemy_imperative 22.7 ms 18.0 ms: 1.26x faster
sympy_sum 193 ms 154 ms: 1.25x faster
fannkuch 483 ms 387 ms: 1.25x faster
json_dumps 14.5 ms 11.8 ms: 1.24x faster
sqlglot_normalize 144 ms 119 ms: 1.21x faster
dulwich_log 81.1 ms 67.1 ms: 1.21x faster
sympy_str 360 ms 298 ms: 1.21x faster
2to3 350 ms 290 ms: 1.21x faster
sympy_integrate 28.2 ms 23.6 ms: 1.19x faster
scimark_fft 361 ms 304 ms: 1.19x faster
xml_etree_parse 160 ms 136 ms: 1.18x faster
sympy_expand 600 ms 512 ms: 1.17x faster
bench_thread_pool 1.14 ms 975 us: 1.17x faster
sqlglot_optimize 70.1 ms 60.0 ms: 1.17x faster
xml_etree_generate 92.3 ms 79.3 ms: 1.16x faster
mdp 3.01 sec 2.59 sec: 1.16x faster
docutils 3.41 sec 2.95 sec: 1.16x faster
nqueens 115 ms 99.6 ms: 1.15x faster
xml_etree_iterparse 110 ms 95.9 ms: 1.15x faster
json_loads 30.3 us 26.6 us: 1.14x faster
genshi_xml 63.3 ms 56.6 ms: 1.12x faster
regex_v8 27.2 ms 25.0 ms: 1.09x faster
regex_dna 261 ms 241 ms: 1.08x faster
sqlite_synth 2.99 us 2.77 us: 1.08x faster
json 5.86 ms 5.48 ms: 1.07x faster
pidigits 271 ms 254 ms: 1.07x faster
scimark_sparse_mat_mult 5.08 ms 4.82 ms: 1.05x faster
meteor_contest 138 ms 133 ms: 1.04x faster
unpack_sequence 59.9 ns 58.3 ns: 1.03x faster
regex_effbot 3.09 ms 3.11 ms: 1.01x slower
asyncio_websockets 390 ms 394 ms: 1.01x slower
async_generators 421 ms 434 ms: 1.03x slower
unpickle_list 4.65 us 4.83 us: 1.04x slower
unpickle 13.5 us 14.2 us: 1.05x slower
telco 7.23 ms 7.59 ms: 1.05x slower
pickle 9.89 us 12.1 us: 1.23x slower
python_startup_no_site 7.33 ms 9.01 ms: 1.23x slower
pickle_dict 29.5 us 37.3 us: 1.26x slower
coverage 63.3 ms 80.8 ms: 1.28x slower
pickle_list 4.12 us 5.36 us: 1.30x slower
python_startup 11.5 ms 16.0 ms: 1.39x slower
create_gc_cycles 1.76 ms 2.67 ms: 1.51x slower
gc_traversal 3.42 ms 6.29 ms: 1.84x slower
bench_mp_pool 6.37 ms 1.23 sec: 193.60x slower
Geometric mean (ref) 1.23x faster
Ignored benchmarks (7) of results/bm-20220323-3.10.4-9d38120/bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120.json: aiohttp, chameleon, dask, flaskblogging, gunicorn, mypy2, tornado_http
Ignored benchmarks (11) of results/bm-20250208-3.14.0a4+-29f8a67-JIT/bm-20250208-pythonperf2-x86_64-python-29f8a67ae00081a36fdc-3.14.0a4+-29f8a67.json: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg, bpe_tokeniser, connected_components, k_core, many_optionals, shortest_path, sphinx, subparsers
  • Geometric mean (including insignificant results): 1.329x faster

HPT report

  • Reliability score: 100.00% likely to be faster
  • 90% likely to have a speedup of 1.25x
  • 95% likely to have a speedup of 1.24x
  • 99% likely to have a speedup of 1.21x

Memory

  • memory change: 1.29x