python性能测试之pyperformance

python性能测试之pyperformance

是用于测试python性能的一个软件套装。

作用是对python不同的版本测试性能，发现Python在实现过程中是否有性能衰减，当然也可以用于测试不同的机器的性能差异。

 官方的文档位于：

Usage — Python Performance Benchmark Suite 1.0.3 documentation

大家注意，大多数情况下，我们只需要看官方文档就可以了。因为其他人写的文档肯定没有官方写的好，写的具体，写的全面。 

我这里只是把大体的情况罗列一下，具体的还是要看官方文档。
 

安装：

python3 -m pip install pyperformance

 运行：  

pyperformance run --python=python3.6 -o py36.json
pyperformance run --python=python3.7 -o py38.json
pyperformance compare py36.json py38.json

 python3 -m pyperformance run -o py38.json

 对结果进行检查：

python3 -m pyperf show py36.json
python3 -m pyperf check py36.json
python3 -m pyperf metadata py36.json
python3 -m pyperf stats py36.json
python3 -m pyperf hist py36.json
python3 -m pyperf dump py36.json

性能对比：

python3 -m pyperf compare_to py36.json py38.json --table

以表格的形式输出的样子：

+-------------------------+----------+------------------------+
| Benchmark               | u540     | u740                   |
+=========================+==========+========================+
| 2to3                    | 8.76 sec | 7.36 sec: 1.19x faster |
+-------------------------+----------+------------------------+
| chameleon               | 369 ms   | 280 ms: 1.32x faster   |
+-------------------------+----------+------------------------+
| chaos                   | 3.18 sec | 2.72 sec: 1.17x faster |
+-------------------------+----------+------------------------+
| crypto_pyaes            | 2.85 sec | 2.47 sec: 1.15x faster |
+-------------------------+----------+------------------------+
| deltablue               | 222 ms   | 186 ms: 1.19x faster   |
+-------------------------+----------+------------------------+
| django_template         | 1.86 sec | 1.62 sec: 1.15x faster |
+-------------------------+----------+------------------------+
| dulwich_log             | 1.63 sec | 1.48 sec: 1.10x faster |
+-------------------------+----------+------------------------+
| fannkuch                | 10.0 sec | 9.46 sec: 1.06x faster |
+-------------------------+----------+------------------------+
| float                   | 3.82 sec | 2.81 sec: 1.36x faster |
+-------------------------+----------+------------------------+
| go                      | 7.28 sec | 6.18 sec: 1.18x faster |
+-------------------------+----------+------------------------+
| hexiom                  | 302 ms   | 230 ms: 1.31x faster   |
+-------------------------+----------+------------------------+
| json_dumps              | 397 ms   | 335 ms: 1.18x faster   |
+-------------------------+----------+------------------------+
| json_loads              | 581 us   | 594 us: 1.02x slower   |
+-------------------------+----------+------------------------+
| logging_format          | 343 us   | 263 us: 1.30x faster   |
+-------------------------+----------+------------------------+
| logging_silent          | 6.65 us  | 4.78 us: 1.39x faster  |
+-------------------------+----------+------------------------+
| logging_simple          | 282 us   | 243 us: 1.16x faster   |
+-------------------------+----------+------------------------+
| mako                    | 581 ms   | 450 ms: 1.29x faster   |
+-------------------------+----------+------------------------+
| meteor_contest          | 2.15 sec | 1.85 sec: 1.16x faster |
+-------------------------+----------+------------------------+
| nbody                   | 2.96 sec | 2.59 sec: 1.14x faster |
+-------------------------+----------+------------------------+
| nqueens                 | 2.73 sec | 2.39 sec: 1.14x faster |
+-------------------------+----------+------------------------+
| pathlib                 | 591 ms   | 517 ms: 1.14x faster   |
+-------------------------+----------+------------------------+
| pickle                  | 200 us   | 217 us: 1.09x slower   |
+-------------------------+----------+------------------------+
| pickle_dict             | 400 us   | 483 us: 1.21x slower   |
+-------------------------+----------+------------------------+
| pickle_list             | 59.0 us  | 65.6 us: 1.11x slower  |
+-------------------------+----------+------------------------+
| pickle_pure_python      | 17.3 ms  | 12.9 ms: 1.35x faster  |
+-------------------------+----------+------------------------+
| pidigits                | 2.36 sec | 3.53 sec: 1.50x slower |
+-------------------------+----------+------------------------+
| pyflate                 | 19.9 sec | 16.6 sec: 1.20x faster |
+-------------------------+----------+------------------------+
| python_startup          | 133 ms   | 121 ms: 1.10x faster   |
+-------------------------+----------+------------------------+
| python_startup_no_site  | 86.7 ms  | 79.0 ms: 1.10x faster  |
+-------------------------+----------+------------------------+
| raytrace                | 15.1 sec | 13.0 sec: 1.17x faster |
+-------------------------+----------+------------------------+
| regex_compile           | 4.90 sec | 3.97 sec: 1.23x faster |
+-------------------------+----------+------------------------+
| regex_dna               | 2.69 sec | 3.15 sec: 1.17x slower |
+-------------------------+----------+------------------------+
| regex_effbot            | 55.6 ms  | 62.7 ms: 1.13x slower  |
+-------------------------+----------+------------------------+
| regex_v8                | 432 ms   | 449 ms: 1.04x slower   |
+-------------------------+----------+------------------------+
| richards                | 2.10 sec | 1.80 sec: 1.17x faster |
+-------------------------+----------+------------------------+
| scimark_fft             | 8.18 sec | 7.24 sec: 1.13x faster |
+-------------------------+----------+------------------------+
| scimark_lu              | 4.96 sec | 3.98 sec: 1.25x faster |
+-------------------------+----------+------------------------+
| scimark_monte_carlo     | 2.94 sec | 2.52 sec: 1.17x faster |
+-------------------------+----------+------------------------+
| scimark_sor             | 5.39 sec | 4.72 sec: 1.14x faster |
+-------------------------+----------+------------------------+
| scimark_sparse_mat_mult | 144 ms   | 101 ms: 1.42x faster   |
+-------------------------+----------+------------------------+
| spectral_norm           | 3.57 sec | 3.18 sec: 1.12x faster |
+-------------------------+----------+------------------------+
| sqlalchemy_declarative  | 3.08 sec | 2.66 sec: 1.16x faster |
+-------------------------+----------+------------------------+
| sqlalchemy_imperative   | 425 ms   | 366 ms: 1.16x faster   |
+-------------------------+----------+------------------------+
| sqlite_synth            | 85.8 us  | 77.2 us: 1.11x faster  |
+-------------------------+----------+------------------------+
| sympy_expand            | 14.5 sec | 12.2 sec: 1.18x faster |
+-------------------------+----------+------------------------+
| sympy_integrate         | 541 ms   | 459 ms: 1.18x faster   |
+-------------------------+----------+------------------------+
| sympy_sum               | 4.07 sec | 3.55 sec: 1.15x faster |
+-------------------------+----------+------------------------+
| sympy_str               | 8.45 sec | 7.08 sec: 1.19x faster |
+-------------------------+----------+------------------------+
| telco                   | 221 ms   | 189 ms: 1.17x faster   |
+-------------------------+----------+------------------------+
| unpack_sequence         | 1.17 us  | 1.05 us: 1.12x faster  |
+-------------------------+----------+------------------------+
| unpickle                | 308 us   | 329 us: 1.07x slower   |
+-------------------------+----------+------------------------+
| unpickle_list           | 76.8 us  | 74.9 us: 1.03x faster  |
+-------------------------+----------+------------------------+
| unpickle_pure_python    | 11.4 ms  | 9.11 ms: 1.25x faster  |
+-------------------------+----------+------------------------+
| xml_etree_parse         | 3.79 sec | 3.41 sec: 1.11x faster |
+-------------------------+----------+------------------------+
| xml_etree_iterparse     | 3.04 sec | 2.48 sec: 1.23x faster |
+-------------------------+----------+------------------------+
| xml_etree_generate      | 3.12 sec | 2.58 sec: 1.21x faster |
+-------------------------+----------+------------------------+
| xml_etree_process       | 2.58 sec | 2.15 sec: 1.20x faster |
+-------------------------+----------+------------------------+
| Geometric mean          | (ref)    | 1.13x faster           |
+-------------------------+----------+------------------------+

 

参考：

Python性能测试之performance | 天问博客