Skip to content

Latest commit

 

History

History
212 lines (192 loc) · 25.4 KB

bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17-vs-3.10.4.md

File metadata and controls

212 lines (192 loc) · 25.4 KB

Results vs. 3.10.4

  • fork: python
  • ref: v3.13.0b2
  • machine: linux-x86_64
  • commit hash: 3a83b17
  • commit date: 2024-06-05
  • overall geometric mean: 1.28x faster
  • HPT reliability: 100.00%
  • HPT 99th percentile: 1.21x faster
  • Memory change: 1.13x

Benchmarks with tag 'apps':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
2to3 350 ms 291 ms: 1.20x faster
chameleon 9.44 ms 7.40 ms: 1.28x faster
docutils 3.41 sec 2.98 sec: 1.14x faster
html5lib 94.6 ms 74.7 ms: 1.27x faster
tornado_http 157 ms 119 ms: 1.31x faster
Geometric mean (ref) 1.24x faster

Benchmarks with tag 'asyncio':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
async_tree_none 692 ms 365 ms: 1.89x faster
async_tree_io 1.60 sec 873 ms: 1.83x faster
async_tree_memoization 819 ms 460 ms: 1.78x faster
async_tree_cpu_io_mixed 936 ms 604 ms: 1.55x faster
Geometric mean (ref) 1.76x faster

Benchmarks with tag 'math':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
nbody 134 ms 87.8 ms: 1.53x faster
float 111 ms 80.2 ms: 1.39x 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-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
regex_compile 190 ms 144 ms: 1.32x faster
regex_dna 261 ms 249 ms: 1.05x faster
regex_v8 27.2 ms 26.0 ms: 1.04x faster
regex_effbot 3.09 ms 3.40 ms: 1.10x slower
Geometric mean (ref) 1.07x faster

Benchmarks with tag 'serialize':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
pickle_pure_python 455 us 307 us: 1.48x faster
unpickle_pure_python 312 us 224 us: 1.39x faster
json_dumps 14.5 ms 10.8 ms: 1.35x faster
xml_etree_process 75.9 ms 59.7 ms: 1.27x faster
tomli_loads 2.92 sec 2.40 sec: 1.21x faster
json_loads 30.3 us 25.0 us: 1.21x faster
xml_etree_parse 160 ms 144 ms: 1.11x faster
xml_etree_iterparse 110 ms 103 ms: 1.08x faster
xml_etree_generate 92.3 ms 85.7 ms: 1.08x faster
unpickle_list 4.65 us 4.70 us: 1.01x slower
pickle 9.89 us 10.6 us: 1.07x slower
pickle_list 4.12 us 4.44 us: 1.08x slower
pickle_dict 29.5 us 32.8 us: 1.11x slower
unpickle 13.5 us 15.7 us: 1.16x slower
Geometric mean (ref) 1.11x faster

Benchmarks with tag 'startup':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
python_startup 11.5 ms 13.2 ms: 1.15x slower
python_startup_no_site 7.33 ms 8.85 ms: 1.21x slower
Geometric mean (ref) 1.18x slower

Benchmarks with tag 'template':

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
mako 14.7 ms 10.4 ms: 1.42x faster
django_template 50.2 ms 39.0 ms: 1.29x faster
genshi_text 31.4 ms 25.9 ms: 1.21x faster
genshi_xml 63.3 ms 58.1 ms: 1.09x faster
Geometric mean (ref) 1.25x faster

All benchmarks:

Benchmark bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120 bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17
typing_runtime_protocols 537 us 171 us: 3.15x faster
deltablue 7.50 ms 3.37 ms: 2.22x faster
asyncio_tcp 779 ms 378 ms: 2.06x faster
asyncio_tcp_ssl 3.10 sec 1.58 sec: 1.96x faster
async_tree_none 692 ms 365 ms: 1.89x faster
raytrace 489 ms 260 ms: 1.88x faster
async_tree_io 1.60 sec 873 ms: 1.83x faster
chaos 109 ms 59.6 ms: 1.82x faster
async_tree_memoization 819 ms 460 ms: 1.78x faster
logging_silent 167 ns 96.3 ns: 1.74x faster
scimark_lu 167 ms 97.5 ms: 1.71x faster
generators 57.3 ms 33.5 ms: 1.71x faster
pylint 566 ms 339 ms: 1.67x faster
scimark_monte_carlo 107 ms 65.5 ms: 1.64x faster
crypto_pyaes 119 ms 73.6 ms: 1.62x faster
sqlglot_parse 2.24 ms 1.39 ms: 1.61x faster
go 262 ms 165 ms: 1.59x faster
comprehensions 26.7 us 17.0 us: 1.57x faster
async_tree_cpu_io_mixed 936 ms 604 ms: 1.55x faster
nbody 134 ms 87.8 ms: 1.53x faster
pyflate 733 ms 482 ms: 1.52x faster
sqlglot_transpile 2.68 ms 1.76 ms: 1.52x faster
scimark_sor 180 ms 119 ms: 1.52x faster
hexiom 9.42 ms 6.35 ms: 1.48x faster
richards_super 90.6 ms 61.2 ms: 1.48x faster
pickle_pure_python 455 us 307 us: 1.48x faster
spectral_norm 139 ms 97.3 ms: 1.43x faster
mako 14.7 ms 10.4 ms: 1.42x faster
richards 75.1 ms 53.4 ms: 1.41x faster
unpickle_pure_python 312 us 224 us: 1.39x faster
logging_simple 8.88 us 6.40 us: 1.39x faster
float 111 ms 80.2 ms: 1.39x faster
coroutines 30.3 ms 22.0 ms: 1.38x faster
pycparser 1.67 sec 1.22 sec: 1.37x faster
fannkuch 483 ms 353 ms: 1.37x faster
logging_format 9.64 us 7.11 us: 1.36x faster
json_dumps 14.5 ms 10.8 ms: 1.35x faster
thrift 1.18 ms 880 us: 1.34x faster
deepcopy_memo 49.8 us 37.3 us: 1.34x faster
regex_compile 190 ms 144 ms: 1.32x faster
tornado_http 157 ms 119 ms: 1.31x faster
nqueens 115 ms 88.4 ms: 1.30x faster
bench_mp_pool 6.37 ms 4.91 ms: 1.30x faster
pprint_pformat 2.15 sec 1.66 sec: 1.30x faster
django_template 50.2 ms 39.0 ms: 1.29x faster
pprint_safe_repr 1.05 sec 818 ms: 1.28x faster
chameleon 9.44 ms 7.40 ms: 1.28x faster
xml_etree_process 75.9 ms 59.7 ms: 1.27x faster
html5lib 94.6 ms 74.7 ms: 1.27x faster
bench_thread_pool 1.14 ms 908 us: 1.26x faster
pathlib 21.4 ms 17.1 ms: 1.25x faster
sympy_sum 193 ms 155 ms: 1.24x faster
deepcopy 468 us 377 us: 1.24x faster
sympy_str 360 ms 295 ms: 1.22x faster
mdp 3.01 sec 2.46 sec: 1.22x faster
mypy2 933 ms 764 ms: 1.22x faster
sympy_integrate 28.2 ms 23.2 ms: 1.22x faster
genshi_text 31.4 ms 25.9 ms: 1.21x faster
tomli_loads 2.92 sec 2.40 sec: 1.21x faster
json_loads 30.3 us 25.0 us: 1.21x faster
dask 472 ms 391 ms: 1.21x faster
dulwich_log 81.1 ms 67.3 ms: 1.21x faster
2to3 350 ms 291 ms: 1.20x faster
sympy_expand 600 ms 501 ms: 1.20x faster
sqlglot_normalize 144 ms 120 ms: 1.20x faster
scimark_sparse_mat_mult 5.08 ms 4.28 ms: 1.19x faster
deepcopy_reduce 4.01 us 3.39 us: 1.18x faster
sqlglot_optimize 70.1 ms 59.5 ms: 1.18x faster
async_generators 421 ms 363 ms: 1.16x faster
scimark_fft 361 ms 312 ms: 1.16x faster
docutils 3.41 sec 2.98 sec: 1.14x faster
xml_etree_parse 160 ms 144 ms: 1.11x faster
gunicorn 1.16 ms 1.04 ms: 1.11x faster
aiohttp 1.19 ms 1.07 ms: 1.11x faster
json 5.86 ms 5.35 ms: 1.10x faster
genshi_xml 63.3 ms 58.1 ms: 1.09x faster
meteor_contest 138 ms 128 ms: 1.08x faster
xml_etree_iterparse 110 ms 103 ms: 1.08x faster
xml_etree_generate 92.3 ms 85.7 ms: 1.08x faster
sqlite_synth 2.99 us 2.80 us: 1.07x faster
pidigits 271 ms 254 ms: 1.07x faster
regex_dna 261 ms 249 ms: 1.05x faster
regex_v8 27.2 ms 26.0 ms: 1.04x faster
unpickle_list 4.65 us 4.70 us: 1.01x slower
asyncio_websockets 390 ms 395 ms: 1.01x slower
flaskblogging 4.39 ms 4.68 ms: 1.07x slower
pickle 9.89 us 10.6 us: 1.07x slower
pickle_list 4.12 us 4.44 us: 1.08x slower
regex_effbot 3.09 ms 3.40 ms: 1.10x slower
pickle_dict 29.5 us 32.8 us: 1.11x slower
create_gc_cycles 1.76 ms 2.00 ms: 1.14x slower
python_startup 11.5 ms 13.2 ms: 1.15x slower
telco 7.23 ms 8.40 ms: 1.16x slower
unpickle 13.5 us 15.7 us: 1.16x slower
python_startup_no_site 7.33 ms 8.85 ms: 1.21x slower
coverage 63.3 ms 83.5 ms: 1.32x slower
gc_traversal 3.42 ms 4.69 ms: 1.37x slower
Geometric mean (ref) 1.28x faster
Ignored benchmarks (3) of results/bm-20220323-3.10.4-9d38120/bm-20220323-pythonperf2-x86_64-python-v3.10.4-3.10.4-9d38120.json: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
Ignored benchmarks (5) of results/bm-20240605-3.13.0b2-3a83b17/bm-20240605-pythonperf2-x86_64-python-v3.13.0b2-3.13.0b2-3a83b17.json: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg, bpe_tokeniser

HPT report

  • Reliability score: 100.00% likely to be faster
  • 90% likely to have a speedup of 1.24x
  • 95% likely to have a speedup of 1.23x
  • 99% likely to have a speedup of 1.21x

Memory

  • memory change: 1.13x