Skip to content

Instantly share code, notes, and snippets.

@KungFuJesus
Created August 24, 2022 17:31
Show Gist options
  • Save KungFuJesus/d649a1a3879b1b2823f92e442c41613c to your computer and use it in GitHub Desktop.
Save KungFuJesus/d649a1a3879b1b2823f92e442c41613c to your computer and use it in GitHub Desktop.
Benchmarks zlib-dougallj vs zlib-ng (realistically compressible)
RUNNING: ./benchmark_zlib_apps_alt --benchmark_repetitions=10 --benchmark_display_aggregates_only=true --benchmark_filter=png_decode_re --benchmark_out=/tmp/tmpi2ugdndb
2022-08-24T13:28:49-04:00
Running ./benchmark_zlib_apps_alt
Run on (28 X 2508.39 MHz CPU s)
CPU Caches:
L1 Data 32 KiB (x14)
L1 Instruction 32 KiB (x14)
L2 Unified 1024 KiB (x14)
L3 Unified 19712 KiB (x1)
Load Average: 0.71, 0.54, 0.37
---------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations
---------------------------------------------------------------------------------------------
png_decode_realistic/png_decode_realistic/0_mean 10780 us 10779 us 10
png_decode_realistic/png_decode_realistic/0_median 10770 us 10769 us 10
png_decode_realistic/png_decode_realistic/0_stddev 43.7 us 43.7 us 10
png_decode_realistic/png_decode_realistic/0_cv 0.41 % 0.41 % 10
png_decode_realistic/png_decode_realistic/1_mean 43447 us 43444 us 10
png_decode_realistic/png_decode_realistic/1_median 43442 us 43439 us 10
png_decode_realistic/png_decode_realistic/1_stddev 34.8 us 34.3 us 10
png_decode_realistic/png_decode_realistic/1_cv 0.08 % 0.08 % 10
png_decode_realistic/png_decode_realistic/2_mean 41877 us 41874 us 10
png_decode_realistic/png_decode_realistic/2_median 41876 us 41874 us 10
png_decode_realistic/png_decode_realistic/2_stddev 61.2 us 60.6 us 10
png_decode_realistic/png_decode_realistic/2_cv 0.15 % 0.14 % 10
png_decode_realistic/png_decode_realistic/3_mean 38986 us 38985 us 10
png_decode_realistic/png_decode_realistic/3_median 38994 us 38993 us 10
png_decode_realistic/png_decode_realistic/3_stddev 27.3 us 27.3 us 10
png_decode_realistic/png_decode_realistic/3_cv 0.07 % 0.07 % 10
png_decode_realistic/png_decode_realistic/4_mean 37741 us 37739 us 10
png_decode_realistic/png_decode_realistic/4_median 37691 us 37690 us 10
png_decode_realistic/png_decode_realistic/4_stddev 159 us 159 us 10
png_decode_realistic/png_decode_realistic/4_cv 0.42 % 0.42 % 10
png_decode_realistic/png_decode_realistic/5_mean 36984 us 36982 us 10
png_decode_realistic/png_decode_realistic/5_median 36976 us 36973 us 10
png_decode_realistic/png_decode_realistic/5_stddev 37.5 us 36.9 us 10
png_decode_realistic/png_decode_realistic/5_cv 0.10 % 0.10 % 10
png_decode_realistic/png_decode_realistic/6_mean 35665 us 35662 us 10
png_decode_realistic/png_decode_realistic/6_median 35666 us 35664 us 10
png_decode_realistic/png_decode_realistic/6_stddev 58.3 us 58.4 us 10
png_decode_realistic/png_decode_realistic/6_cv 0.16 % 0.16 % 10
png_decode_realistic/png_decode_realistic/7_mean 35349 us 35348 us 10
png_decode_realistic/png_decode_realistic/7_median 35343 us 35342 us 10
png_decode_realistic/png_decode_realistic/7_stddev 49.6 us 49.6 us 10
png_decode_realistic/png_decode_realistic/7_cv 0.14 % 0.14 % 10
png_decode_realistic/png_decode_realistic/8_mean 34295 us 34294 us 10
png_decode_realistic/png_decode_realistic/8_median 34285 us 34284 us 10
png_decode_realistic/png_decode_realistic/8_stddev 70.6 us 70.6 us 10
png_decode_realistic/png_decode_realistic/8_cv 0.21 % 0.21 % 10
png_decode_realistic/png_decode_realistic/9_mean 33617 us 33615 us 10
png_decode_realistic/png_decode_realistic/9_median 33634 us 33631 us 10
png_decode_realistic/png_decode_realistic/9_stddev 55.2 us 54.8 us 10
png_decode_realistic/png_decode_realistic/9_cv 0.16 % 0.16 % 10
RUNNING: ./benchmark_zlib_apps --benchmark_repetitions=10 --benchmark_display_aggregates_only=true --benchmark_filter=png_decode_re --benchmark_out=/tmp/tmpg2bd89i6
2022-08-24T13:30:01-04:00
Running ./benchmark_zlib_apps
Run on (28 X 4101.09 MHz CPU s)
CPU Caches:
L1 Data 32 KiB (x14)
L1 Instruction 32 KiB (x14)
L2 Unified 1024 KiB (x14)
L3 Unified 19712 KiB (x1)
Load Average: 1.08, 0.69, 0.43
---------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations
---------------------------------------------------------------------------------------------
png_decode_realistic/png_decode_realistic/0_mean 9853 us 9853 us 10
png_decode_realistic/png_decode_realistic/0_median 9861 us 9861 us 10
png_decode_realistic/png_decode_realistic/0_stddev 24.2 us 24.2 us 10
png_decode_realistic/png_decode_realistic/0_cv 0.25 % 0.25 % 10
png_decode_realistic/png_decode_realistic/1_mean 58802 us 58799 us 10
png_decode_realistic/png_decode_realistic/1_median 58450 us 58448 us 10
png_decode_realistic/png_decode_realistic/1_stddev 633 us 633 us 10
png_decode_realistic/png_decode_realistic/1_cv 1.08 % 1.08 % 10
png_decode_realistic/png_decode_realistic/2_mean 55574 us 55572 us 10
png_decode_realistic/png_decode_realistic/2_median 55547 us 55545 us 10
png_decode_realistic/png_decode_realistic/2_stddev 321 us 321 us 10
png_decode_realistic/png_decode_realistic/2_cv 0.58 % 0.58 % 10
png_decode_realistic/png_decode_realistic/3_mean 50502 us 50499 us 10
png_decode_realistic/png_decode_realistic/3_median 50485 us 50482 us 10
png_decode_realistic/png_decode_realistic/3_stddev 339 us 339 us 10
png_decode_realistic/png_decode_realistic/3_cv 0.67 % 0.67 % 10
png_decode_realistic/png_decode_realistic/4_mean 46905 us 46902 us 10
png_decode_realistic/png_decode_realistic/4_median 46874 us 46872 us 10
png_decode_realistic/png_decode_realistic/4_stddev 275 us 274 us 10
png_decode_realistic/png_decode_realistic/4_cv 0.59 % 0.59 % 10
png_decode_realistic/png_decode_realistic/5_mean 45560 us 45557 us 10
png_decode_realistic/png_decode_realistic/5_median 45612 us 45609 us 10
png_decode_realistic/png_decode_realistic/5_stddev 245 us 245 us 10
png_decode_realistic/png_decode_realistic/5_cv 0.54 % 0.54 % 10
png_decode_realistic/png_decode_realistic/6_mean 43324 us 43322 us 10
png_decode_realistic/png_decode_realistic/6_median 43233 us 43231 us 10
png_decode_realistic/png_decode_realistic/6_stddev 300 us 300 us 10
png_decode_realistic/png_decode_realistic/6_cv 0.69 % 0.69 % 10
png_decode_realistic/png_decode_realistic/7_mean 42595 us 42594 us 10
png_decode_realistic/png_decode_realistic/7_median 42619 us 42618 us 10
png_decode_realistic/png_decode_realistic/7_stddev 301 us 301 us 10
png_decode_realistic/png_decode_realistic/7_cv 0.71 % 0.71 % 10
png_decode_realistic/png_decode_realistic/8_mean 41391 us 41390 us 10
png_decode_realistic/png_decode_realistic/8_median 41352 us 41351 us 10
png_decode_realistic/png_decode_realistic/8_stddev 224 us 224 us 10
png_decode_realistic/png_decode_realistic/8_cv 0.54 % 0.54 % 10
png_decode_realistic/png_decode_realistic/9_mean 40741 us 40740 us 10
png_decode_realistic/png_decode_realistic/9_median 40676 us 40675 us 10
png_decode_realistic/png_decode_realistic/9_stddev 206 us 206 us 10
png_decode_realistic/png_decode_realistic/9_cv 0.51 % 0.51 % 10
Comparing png_decode_re (from ./benchmark_zlib_apps_alt) to png_decode_re (from ./benchmark_zlib_apps)
Benchmark Time CPU Time Old Time New CPU Old CPU New
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/0_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/0_mean -0.0860 -0.0860 10780 9853 10779 9853
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/0_median -0.0844 -0.0844 10770 9861 10769 9861
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/0_stddev -0.4470 -0.4465 44 24 44 24
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/0_cv -0.3950 -0.3945 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/1_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/1_mean +0.3534 +0.3534 43447 58802 43444 58799
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/1_median +0.3454 +0.3455 43442 58450 43439 58448
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/1_stddev +17.2045 +17.4790 35 633 34 633
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/1_cv +12.4506 +12.6535 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/2_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/2_mean +0.3271 +0.3271 41877 55574 41874 55572
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/2_median +0.3264 +0.3265 41876 55547 41874 55545
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/2_stddev +4.2384 +4.2946 61 321 61 321
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/2_cv +2.9472 +2.9896 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/3_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/3_mean +0.2954 +0.2953 38986 50502 38985 50499
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/3_median +0.2947 +0.2946 38994 50485 38993 50482
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/3_stddev +11.3882 +11.4331 27 339 27 339
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/3_cv +8.5633 +8.5984 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/4_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/4_mean +0.2428 +0.2428 37741 46905 37739 46902
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/4_median +0.2436 +0.2436 37691 46874 37690 46872
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/4_stddev +0.7245 +0.7251 159 275 159 274
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/4_cv +0.3876 +0.3881 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/5_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/5_mean +0.2319 +0.2319 36984 45560 36982 45557
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/5_median +0.2336 +0.2336 36976 45612 36973 45609
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/5_stddev +5.5453 +5.6177 37 245 37 245
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/5_cv +4.3132 +4.3720 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/6_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/6_mean +0.2148 +0.2148 35665 43324 35662 43322
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/6_median +0.2121 +0.2122 35666 43233 35664 43231
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/6_stddev +4.1487 +4.1409 58 300 58 300
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/6_cv +3.2385 +3.2320 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/7_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/7_mean +0.2050 +0.2050 35349 42595 35348 42594
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/7_median +0.2059 +0.2059 35343 42619 35342 42618
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/7_stddev +5.0635 +5.0647 50 301 50 301
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/7_cv +4.0320 +4.0330 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/8_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/8_mean +0.2069 +0.2069 34295 41391 34294 41390
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/8_median +0.2061 +0.2061 34285 41352 34284 41351
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/8_stddev +2.1774 +2.1734 71 224 71 224
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/8_cv +1.6326 +1.6293 0 0 0 0
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/9_pvalue 0.0002 0.0002 U Test, Repetitions: 10 vs 10
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/9_mean +0.2119 +0.2119 33617 40741 33615 40740
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/9_median +0.2094 +0.2094 33634 40676 33631 40675
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/9_stddev +2.7263 +2.7525 55 206 55 206
[png_decode_re vs. png_decode_re]alistic/[png_decode_re vs. png_decode_re]alistic/9_cv +2.0746 +2.0962 0 0 0 0
OVERALL_GEOMEAN +0.2143 +0.2143 0 0 0 0
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment