Study
[Python] Version 3.11 Release Review
chn1002
2022. 5. 19. 13:52
Python 3.11 RR
Python 3.11 이 2022. 5월 18일 RR 되었다. Python 3.11 은 기존 Python 3.10 보다 최대 10 ~ 60% 빠르다. 평균적으로 표준 벤치마킹 제품군에서 1.25배의 속도 향상을 보인다.
- CPython 3.11 은 pyperformance 벤치마킹
| Benchmark | 2022-05-05_17-47-3.10-9b47252d5433 | 2022-05-05_16-04-main-ae553b3561f0 |
| deltablue | 12.4 ms | 6.35 ms: 1.96x faster |
| go | 381 ms | 229 ms: 1.66x faster |
| logging_silent | 281 ns | 169 ns: 1.66x faster |
| scimark_sor | 324 ms | 199 ms: 1.63x faster |
| scimark_monte_carlo | 177 ms | 113 ms: 1.57x faster |
| raytrace | 782 ms | 505 ms: 1.55x faster |
| richards | 123 ms | 79.5 ms: 1.54x faster |
| pyflate | 1.08 sec | 699 ms: 1.54x faster |
| crypto_pyaes | 193 ms | 126 ms: 1.54x faster |
| chaos | 174 ms | 114 ms: 1.53x faster |
| float | 184 ms | 125 ms: 1.48x faster |
| pickle_pure_python | 740 us | 502 us: 1.48x faster |
| scimark_lu | 274 ms | 187 ms: 1.47x faster |
| nbody | 229 ms | 158 ms: 1.45x faster |
| hexiom | 15.4 ms | 10.6 ms: 1.45x faster |
| mako | 25.1 ms | 17.6 ms: 1.43x faster |
| spectral_norm | 248 ms | 174 ms: 1.43x faster |
| xml_etree_process | 125 ms | 94.4 ms: 1.33x faster |
| logging_simple | 13.5 us | 10.2 us: 1.33x faster |
| django_template | 73.5 ms | 56.0 ms: 1.31x faster |
| logging_format | 14.5 us | 11.0 us: 1.31x faster |
| unpickle_pure_python | 490 us | 382 us: 1.28x faster |
| tornado_http | 212 ms | 165 ms: 1.28x faster |
| regex_compile | 286 ms | 223 ms: 1.28x faster |
| 2to3 | 533 ms | 419 ms: 1.27x faster |
| chameleon | 14.5 ms | 11.6 ms: 1.25x faster |
| fannkuch | 784 ms | 642 ms: 1.22x faster |
| sqlalchemy_imperative | 35.0 ms | 29.1 ms: 1.20x faster |
| sqlalchemy_declarative | 267 ms | 222 ms: 1.20x faster |
| scimark_fft | 677 ms | 568 ms: 1.19x faster |
| scimark_sparse_mat_mult | 8.87 ms | 7.46 ms: 1.19x faster |
| sqlite_synth | 4.80 us | 4.06 us: 1.18x faster |
| dulwich_log | 125 ms | 106 ms: 1.18x faster |
| nqueens | 164 ms | 139 ms: 1.18x faster |
| xml_etree_generate | 157 ms | 134 ms: 1.17x faster |
| unpack_sequence | 96.9 ns | 82.9 ns: 1.17x faster |
| sympy_integrate | 39.2 ms | 34.2 ms: 1.15x faster |
| sympy_expand | 869 ms | 772 ms: 1.13x faster |
| sympy_sum | 303 ms | 270 ms: 1.12x faster |
| sympy_str | 527 ms | 472 ms: 1.12x faster |
| regex_v8 | 39.7 ms | 35.8 ms: 1.11x faster |
| regex_effbot | 5.30 ms | 4.79 ms: 1.10x faster |
| python_startup | 15.4 ms | 13.9 ms: 1.10x faster |
| json_loads | 48.5 us | 44.1 us: 1.10x faster |
| pathlib | 33.6 ms | 30.8 ms: 1.09x faster |
| unpickle | 23.7 us | 21.9 us: 1.08x faster |
| pickle_list | 7.55 us | 7.04 us: 1.07x faster |
| json_dumps | 22.6 ms | 21.3 ms: 1.06x faster |
| xml_etree_iterparse | 176 ms | 167 ms: 1.05x faster |
| regex_dna | 353 ms | 337 ms: 1.05x faster |
| meteor_contest | 192 ms | 185 ms: 1.04x faster |
| pickle | 17.2 us | 16.7 us: 1.03x faster |
| xml_etree_parse | 255 ms | 248 ms: 1.03x faster |
| telco | 11.4 ms | 11.1 ms: 1.02x faster |
| pickle_dict | 50.4 us | 49.6 us: 1.02x faster |
| python_startup_no_site | 9.98 ms | 10.4 ms: 1.04x slower |
| pidigits | 316 ms | 346 ms: 1.10x slower |
| Geometric mean | (ref) | 1.25x faster |
- 새로운 문법
- PEP 654 : 예외 그룹 및except*. (Bpo -45292의 Irit Katriel 제공 )
- 새로운 Typing 기능
- 보안 개선 사항
- 새로운 -P명령줄 옵션 및PYTHONSAFEPATHsys.path현재 디렉토리, 스크립트의 디렉토리 또는 빈 문자열과 같이 잠재적으로 안전하지 않은 경로를 앞에 추가하지 않도록 환경 변수를 지정.