Skip to content

Latest commit

 

History

History
205 lines (182 loc) · 25 KB

bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4-vs-3.13.0rc2.md

File metadata and controls

205 lines (182 loc) · 25 KB

Results vs. 3.13.0rc2

  • fork: corona10
  • ref: gh_115999_BINARY_SUB
  • machine: linux-x86_64
  • commit hash: 47b80b4
  • commit date: 2024-12-19
  • overall geometric mean: 1.246x slower
  • HPT reliability: 100.00%
  • HPT 99th percentile: 1.19x slower
  • Memory change: 1.31x

Benchmarks with tag 'apps':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
2to3 260 ms 366 ms: 1.41x slower
docutils 2.62 sec 3.10 sec: 1.18x slower
html5lib 67.0 ms 98.8 ms: 1.47x slower
Geometric mean (ref) 1.35x slower

Benchmarks with tag 'asyncio':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
async_tree_io_tg 913 ms 823 ms: 1.11x faster
async_tree_io 876 ms 852 ms: 1.03x faster
async_tree_cpu_io_mixed 666 ms 652 ms: 1.02x faster
asyncio_websockets 520 ms 517 ms: 1.01x faster
coroutines 23.6 ms 24.6 ms: 1.04x slower
async_tree_none_tg 336 ms 358 ms: 1.06x slower
async_tree_memoization_tg 414 ms 443 ms: 1.07x slower
async_tree_none 354 ms 387 ms: 1.09x slower
async_generators 377 ms 452 ms: 1.20x slower
Geometric mean (ref) 1.03x slower

Benchmark hidden because not significant (2): async_tree_cpu_io_mixed_tg, async_tree_memoization

Benchmarks with tag 'math':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
pidigits 217 ms 182 ms: 1.19x faster
nbody 85.1 ms 134 ms: 1.57x slower
float 77.5 ms 140 ms: 1.81x slower
Geometric mean (ref) 1.34x slower

Benchmarks with tag 'regex':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
regex_effbot 3.08 ms 2.90 ms: 1.06x faster
regex_dna 180 ms 184 ms: 1.02x slower
regex_v8 22.7 ms 24.9 ms: 1.10x slower
regex_compile 132 ms 179 ms: 1.35x slower
Geometric mean (ref) 1.09x slower

Benchmarks with tag 'serialize':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
xml_etree_iterparse 94.9 ms 91.1 ms: 1.04x faster
xml_etree_parse 136 ms 131 ms: 1.04x faster
json_loads 27.0 us 28.2 us: 1.04x slower
xml_etree_generate 85.4 ms 97.3 ms: 1.14x slower
xml_etree_process 59.3 ms 77.4 ms: 1.30x slower
tomli_loads 2.01 sec 2.66 sec: 1.33x slower
json_dumps 10.5 ms 14.5 ms: 1.37x slower
unpickle_pure_python 210 us 335 us: 1.60x slower
pickle_pure_python 294 us 511 us: 1.73x slower
Geometric mean (ref) 1.25x slower

Benchmarks with tag 'startup':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
python_startup_no_site 7.39 ms 10.2 ms: 1.38x slower
python_startup 11.0 ms 17.4 ms: 1.59x slower
Geometric mean (ref) 1.48x slower

Benchmarks with tag 'template':

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
genshi_xml 48.8 ms 63.3 ms: 1.30x slower
genshi_text 21.5 ms 31.7 ms: 1.47x slower
django_template 34.1 ms 51.4 ms: 1.51x slower
mako 11.3 ms 17.7 ms: 1.56x slower
Geometric mean (ref) 1.46x slower

All benchmarks:

Benchmark bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006 bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4
pidigits 217 ms 182 ms: 1.19x faster
async_tree_io_tg 913 ms 823 ms: 1.11x faster
deepcopy 355 us 327 us: 1.09x faster
regex_effbot 3.08 ms 2.90 ms: 1.06x faster
xml_etree_iterparse 94.9 ms 91.1 ms: 1.04x faster
xml_etree_parse 136 ms 131 ms: 1.04x faster
async_tree_io 876 ms 852 ms: 1.03x faster
async_tree_cpu_io_mixed 666 ms 652 ms: 1.02x faster
asyncio_websockets 520 ms 517 ms: 1.01x faster
gc_traversal 3.14 ms 3.17 ms: 1.01x slower
deepcopy_memo 39.1 us 39.5 us: 1.01x slower
regex_dna 180 ms 184 ms: 1.02x slower
json 4.93 ms 5.11 ms: 1.04x slower
sqlite_synth 2.21 us 2.29 us: 1.04x slower
json_loads 27.0 us 28.2 us: 1.04x slower
coroutines 23.6 ms 24.6 ms: 1.04x slower
async_tree_none_tg 336 ms 358 ms: 1.06x slower
pathlib 19.2 ms 20.4 ms: 1.07x slower
async_tree_memoization_tg 414 ms 443 ms: 1.07x slower
spectral_norm 111 ms 120 ms: 1.08x slower
async_tree_none 354 ms 387 ms: 1.09x slower
regex_v8 22.7 ms 24.9 ms: 1.10x slower
telco 7.82 ms 8.66 ms: 1.11x slower
deepcopy_reduce 3.11 us 3.48 us: 1.12x slower
bpe_tokeniser 4.45 sec 5.00 sec: 1.13x slower
xml_etree_generate 85.4 ms 97.3 ms: 1.14x slower
scimark_fft 349 ms 402 ms: 1.15x slower
docutils 2.62 sec 3.10 sec: 1.18x slower
coverage 83.0 ms 99.1 ms: 1.19x slower
async_generators 377 ms 452 ms: 1.20x slower
pylint 317 ms 381 ms: 1.20x slower
mdp 2.36 sec 2.84 sec: 1.21x slower
nqueens 78.6 ms 96.0 ms: 1.22x slower
scimark_sparse_mat_mult 4.71 ms 5.93 ms: 1.26x slower
meteor_contest 102 ms 128 ms: 1.26x slower
dulwich_log 74.8 ms 95.5 ms: 1.28x slower
sqlglot_normalize 106 ms 135 ms: 1.28x slower
genshi_xml 48.8 ms 63.3 ms: 1.30x slower
xml_etree_process 59.3 ms 77.4 ms: 1.30x slower
sqlglot_optimize 52.7 ms 68.8 ms: 1.31x slower
pprint_safe_repr 738 ms 970 ms: 1.32x slower
typing_runtime_protocols 155 us 204 us: 1.32x slower
generators 28.8 ms 38.1 ms: 1.32x slower
tomli_loads 2.01 sec 2.66 sec: 1.33x slower
fannkuch 370 ms 492 ms: 1.33x slower
create_gc_cycles 1.34 ms 1.80 ms: 1.35x slower
pprint_pformat 1.50 sec 2.02 sec: 1.35x slower
regex_compile 132 ms 179 ms: 1.35x slower
crypto_pyaes 67.9 ms 92.1 ms: 1.36x slower
pycparser 1.12 sec 1.52 sec: 1.37x slower
json_dumps 10.5 ms 14.5 ms: 1.37x slower
python_startup_no_site 7.39 ms 10.2 ms: 1.38x slower
scimark_lu 113 ms 156 ms: 1.39x slower
2to3 260 ms 366 ms: 1.41x slower
genshi_text 21.5 ms 31.7 ms: 1.47x slower
thrift 778 us 1.15 ms: 1.47x slower
html5lib 67.0 ms 98.8 ms: 1.47x slower
sympy_integrate 19.8 ms 29.4 ms: 1.48x slower
django_template 34.1 ms 51.4 ms: 1.51x slower
pyflate 449 ms 691 ms: 1.54x slower
mako 11.3 ms 17.7 ms: 1.56x slower
nbody 85.1 ms 134 ms: 1.57x slower
python_startup 11.0 ms 17.4 ms: 1.59x slower
unpickle_pure_python 210 us 335 us: 1.60x slower
hexiom 5.99 ms 9.72 ms: 1.62x slower
scimark_sor 134 ms 221 ms: 1.64x slower
comprehensions 16.5 us 27.5 us: 1.67x slower
richards_super 51.6 ms 86.9 ms: 1.68x slower
richards 45.2 ms 76.9 ms: 1.70x slower
logging_format 6.84 us 11.8 us: 1.72x slower
logging_simple 6.16 us 10.6 us: 1.72x slower
sympy_str 275 ms 476 ms: 1.73x slower
pickle_pure_python 294 us 511 us: 1.73x slower
logging_silent 103 ns 184 ns: 1.79x slower
float 77.5 ms 140 ms: 1.81x slower
chaos 57.3 ms 105 ms: 1.83x slower
scimark_monte_carlo 65.4 ms 120 ms: 1.84x slower
sqlglot_transpile 1.56 ms 2.97 ms: 1.91x slower
go 141 ms 271 ms: 1.92x slower
sqlglot_parse 1.25 ms 2.57 ms: 2.06x slower
sympy_expand 457 ms 953 ms: 2.08x slower
raytrace 253 ms 547 ms: 2.17x slower
sympy_sum 156 ms 350 ms: 2.25x slower
deltablue 3.12 ms 7.98 ms: 2.55x slower
bench_thread_pool 919 us 3.43 ms: 3.73x slower
bench_mp_pool 11.0 ms 109 ms: 9.87x slower
Geometric mean (ref) 1.37x slower

Benchmark hidden because not significant (2): async_tree_cpu_io_mixed_tg, async_tree_memoization Ignored benchmarks (14) of results/bm-20240906-3.13.0rc2-ec61006/bm-20240906-vultr-x86_64-python-v3.13.0rc2-3.13.0rc2-ec61006.json: aiohttp, asyncio_tcp, asyncio_tcp_ssl, chameleon, dask, flaskblogging, gunicorn, pickle, pickle_dict, pickle_list, tornado_http, unpack_sequence, unpickle, unpickle_list Ignored benchmarks (8) of results/bm-20241219-3.14.0a2+-47b80b4-NOGIL/bm-20241219-vultr-x86_64-corona10-gh_115999_BINARY_SUB-3.14.0a2+-47b80b4.json: connected_components, k_core, many_optionals, shortest_path, sphinx, sqlalchemy_declarative, sqlalchemy_imperative, subparsers

  • Geometric mean (including insignificant results): 1.246x slower

HPT report

  • Reliability score: 100.00% likely to be slow
  • 90% likely to have a slowdown of 1.23x
  • 95% likely to have a slowdown of 1.21x
  • 99% likely to have a slowdown of 1.19x

Memory

  • memory change: 1.31x