-
Notifications
You must be signed in to change notification settings - Fork 34
Benchmarks
# JMH version: 1.21
# VM version: JDK 1.8.0_172, Java HotSpot(TM) 64-Bit Server VM, 25.172-b11
# Warmup: 3 iterations, 5 s each
# Measurement: 5 iterations, 10 s each
# Timeout: 10 min per iteration
# Threads: 1 thread, will synchronize iterations
# Benchmark mode: Throughput, ops/time
# i7 7700K / 24G
EncodingJmhBenchmark.encodeBase64Apache 1 thrpt 5 1318131,937 ± 16339,198 ops/s
EncodingJmhBenchmark.encodeBase64Apache 16 thrpt 5 1069180,566 ± 1174,315 ops/s
EncodingJmhBenchmark.encodeBase64Apache 32 thrpt 5 952877,557 ± 1034,081 ops/s
EncodingJmhBenchmark.encodeBase64Apache 128 thrpt 5 470671,111 ± 3890,514 ops/s
EncodingJmhBenchmark.encodeBase64Apache 512 thrpt 5 172110,829 ± 213,959 ops/s
EncodingJmhBenchmark.encodeBase64Apache 1000000 thrpt 5 102,850 ± 3,624 ops/s
EncodingJmhBenchmark.encodeBase64Guava 1 thrpt 5 10807902,000 ± 909701,319 ops/s
EncodingJmhBenchmark.encodeBase64Guava 16 thrpt 5 5703135,415 ± 7063,849 ops/s
EncodingJmhBenchmark.encodeBase64Guava 32 thrpt 5 3875269,919 ± 5242,293 ops/s
EncodingJmhBenchmark.encodeBase64Guava 128 thrpt 5 1378103,987 ± 1333,459 ops/s
EncodingJmhBenchmark.encodeBase64Guava 512 thrpt 5 346530,050 ± 407,467 ops/s
EncodingJmhBenchmark.encodeBase64Guava 1000000 thrpt 5 232,082 ± 4,882 ops/s
EncodingJmhBenchmark.encodeBase64Okio 1 thrpt 5 14068294,672 ± 3751801,084 ops/s
EncodingJmhBenchmark.encodeBase64Okio 16 thrpt 5 6973717,092 ± 97091,658 ops/s
EncodingJmhBenchmark.encodeBase64Okio 32 thrpt 5 4641793,472 ± 107239,983 ops/s
EncodingJmhBenchmark.encodeBase64Okio 128 thrpt 5 1771447,895 ± 28595,096 ops/s
EncodingJmhBenchmark.encodeBase64Okio 512 thrpt 5 276164,010 ± 6808,837 ops/s
EncodingJmhBenchmark.encodeBase64Okio 1000000 thrpt 5 102,610 ± 5,899 ops/s
EncodingJmhBenchmark.encodeBase64OpenJdk 1 thrpt 5 11713907,451 ± 1928162,113 ops/s
EncodingJmhBenchmark.encodeBase64OpenJdk 16 thrpt 5 6816928,904 ± 30601,543 ops/s
EncodingJmhBenchmark.encodeBase64OpenJdk 32 thrpt 5 5087048,151 ± 14096,769 ops/s
EncodingJmhBenchmark.encodeBase64OpenJdk 128 thrpt 5 1905604,168 ± 40709,235 ops/s
EncodingJmhBenchmark.encodeBase64OpenJdk 512 thrpt 5 568632,043 ± 428,865 ops/s
EncodingJmhBenchmark.encodeBase64OpenJdk 1000000 thrpt 5 270,155 ± 29,743 ops/s
Benchmark (byteLength) Mode Cnt Score Error Units
EncodingJmhBenchmark.decodeOnlyBase64Okio 512 thrpt 3 275981,072 ± 4374,638 ops/s
EncodingJmhBenchmark.decodeOnlyBase64Okio 1000000 thrpt 3 130,277 ± 5,843 ops/s
EncodingJmhBenchmark.decodeOnlyBase64OpenJdk 512 thrpt 3 920051,263 ± 17679,192 ops/s
EncodingJmhBenchmark.decodeOnlyBase64OpenJdk 1000000 thrpt 3 424,115 ± 8,624 ops/s
EncodingJmhBenchmark.encodeOnlyBase64Okio 512 thrpt 3 973920,818 ± 23511,284 ops/s
EncodingJmhBenchmark.encodeOnlyBase64Okio 1000000 thrpt 3 485,362 ± 10,893 ops/s
EncodingJmhBenchmark.encodeOnlyBase64OpenJdk 512 thrpt 3 1345013,354 ± 49538,081 ops/s
EncodingJmhBenchmark.encodeOnlyBase64OpenJdk 1000000 thrpt 3 751,239 ± 64,040 ops/s
Result "at.favre.lib.bytes.EncodingHexJmhBenchmark.encodeStackOverflowCode2":
1122,293 ±(99.9%) 22,228 ops/s [Average]
(min, avg, max) = (1120,894, 1122,293, 1123,116), stdev = 1,218
CI (99.9%): [1100,065, 1144,521] (assumes normal distribution)
# Run complete. Total time: 00:25:40
REMEMBER: The numbers below are just data. To gain reusable insights, you need to follow up on
why the numbers are the way they are. Use profilers (see -prof, -lprof), design factorial
experiments, perform baseline and negative tests that provide experimental control, make sure
the benchmarking environment is safe on JVM/OS/HW level, ask for reviews from the domain experts.
Do not assume the numbers tell you what you want them to tell.
NOTE: Current JVM experimentally supports Compiler Blackholes, and they are in use. Please exercise
extra caution when trusting the results, look into the generated code to check the benchmark still
works, and factor in a small probability of new VM bugs. Additionally, while comparisons between
different JVMs are already problematic, the performance difference caused by different Blackhole
modes can be very significant. Please make sure you use the consistent Blackhole mode for comparisons.
Benchmark (byteLength) Mode Cnt Score Error Units
EncodingHexJmhBenchmark.encodeBigInteger 4 thrpt 3 13428556,267 ± 2485252,488 ops/s
EncodingHexJmhBenchmark.encodeBigInteger 8 thrpt 3 4944548,258 ± 36914,459 ops/s
EncodingHexJmhBenchmark.encodeBigInteger 16 thrpt 3 2941403,105 ± 53464,757 ops/s
EncodingHexJmhBenchmark.encodeBigInteger 32 thrpt 3 1389448,671 ± 133105,711 ops/s
EncodingHexJmhBenchmark.encodeBigInteger 128 thrpt 3 242096,448 ± 4847,126 ops/s
EncodingHexJmhBenchmark.encodeBigInteger 512 thrpt 3 43931,400 ± 408,381 ops/s
EncodingHexJmhBenchmark.encodeBigInteger 1000000 thrpt 3 5,335 ± 0,068 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 4 thrpt 3 49699573,884 ± 718083,037 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 8 thrpt 3 39837245,684 ± 194783,743 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 16 thrpt 3 31724981,604 ± 244938,749 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 32 thrpt 3 22786906,028 ± 1107024,474 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 128 thrpt 3 6197028,261 ± 704040,867 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 512 thrpt 3 1504695,048 ± 24497,811 ops/s
EncodingHexJmhBenchmark.encodeBytesLib 1000000 thrpt 3 930,265 ± 130,086 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 4 thrpt 3 23410373,676 ± 7239888,483 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 8 thrpt 3 19364059,826 ± 29244077,987 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 16 thrpt 3 16227097,506 ± 424439,581 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 32 thrpt 3 7768193,947 ± 132317,714 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 128 thrpt 3 3279449,595 ± 34562,168 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 512 thrpt 3 929069,735 ± 138616,676 ops/s
EncodingHexJmhBenchmark.encodeOldBytesLib 1000000 thrpt 3 138,958 ± 0,908 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 4 thrpt 3 48166926,547 ± 3266017,792 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 8 thrpt 3 39284153,920 ± 11076984,360 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 16 thrpt 3 31889491,493 ± 2807579,608 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 32 thrpt 3 22880701,641 ± 2424083,371 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 128 thrpt 3 7677099,242 ± 2594554,970 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 512 thrpt 3 1862006,524 ± 71913,256 ops/s
EncodingHexJmhBenchmark.encodeSchallinsky 1000000 thrpt 3 802,081 ± 175,500 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 4 thrpt 3 47413050,123 ± 2725992,565 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 8 thrpt 3 40677495,172 ± 2817227,931 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 16 thrpt 3 32959351,890 ± 3106728,795 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 32 thrpt 3 23611836,744 ± 877589,728 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 128 thrpt 3 7882848,672 ± 447275,782 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 512 thrpt 3 1882316,791 ± 36826,620 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode1 1000000 thrpt 3 949,783 ± 18,475 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 4 thrpt 3 51237081,006 ± 6255497,646 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 8 thrpt 3 42742454,781 ± 5897169,657 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 16 thrpt 3 34528484,827 ± 1548421,555 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 32 thrpt 3 25978276,665 ± 3060288,487 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 128 thrpt 3 9887505,224 ± 10028617,360 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 512 thrpt 3 2250028,032 ± 386443,795 ops/s
EncodingHexJmhBenchmark.encodeStackOverflowCode2 1000000 thrpt 3 1122,293 ± 22,228 ops/s
Result "at.favre.lib.bytes.EncodingHexJmhBenchmark.encodeStackOverflowCode2": 1049,595 ±(99.9%) 93,412 ops/s [Average] (min, avg, max) = (1045,195, 1049,595, 1055,215), stdev = 5,120 CI (99.9%): [956,183, 1143,007] (assumes normal distribution)
REMEMBER: The numbers below are just data. To gain reusable insights, you need to follow up on why the numbers are the way they are. Use profilers (see -prof, -lprof), design factorial experiments, perform baseline and negative tests that provide experimental control, make sure the benchmarking environment is safe on JVM/OS/HW level, ask for reviews from the domain experts. Do not assume the numbers tell you what you want them to tell.
Benchmark (byteLength) Mode Cnt Score Error Units EncodingHexJmhBenchmark.encodeBigInteger 4 thrpt 3 9333868,849 ± 3894397,371 ops/s EncodingHexJmhBenchmark.encodeBigInteger 8 thrpt 3 4047220,801 ± 429446,689 ops/s EncodingHexJmhBenchmark.encodeBigInteger 16 thrpt 3 2407526,211 ± 114239,702 ops/s EncodingHexJmhBenchmark.encodeBigInteger 32 thrpt 3 1220847,723 ± 42102,349 ops/s EncodingHexJmhBenchmark.encodeBigInteger 128 thrpt 3 211012,498 ± 8304,562 ops/s EncodingHexJmhBenchmark.encodeBigInteger 512 thrpt 3 40888,638 ± 1577,013 ops/s EncodingHexJmhBenchmark.encodeBigInteger 1000000 thrpt 3 5,401 ± 0,149 ops/s EncodingHexJmhBenchmark.encodeBytesLib 4 thrpt 3 30613483,876 ± 537736,621 ops/s EncodingHexJmhBenchmark.encodeBytesLib 8 thrpt 3 21432227,942 ± 2686923,886 ops/s EncodingHexJmhBenchmark.encodeBytesLib 16 thrpt 3 19390972,849 ± 665525,732 ops/s EncodingHexJmhBenchmark.encodeBytesLib 32 thrpt 3 15213485,653 ± 729118,879 ops/s EncodingHexJmhBenchmark.encodeBytesLib 128 thrpt 3 6591738,552 ± 197300,348 ops/s EncodingHexJmhBenchmark.encodeBytesLib 512 thrpt 3 1772463,045 ± 457442,875 ops/s EncodingHexJmhBenchmark.encodeBytesLib 1000000 thrpt 3 896,453 ± 29,679 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 4 thrpt 3 20947931,772 ± 23857593,063 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 8 thrpt 3 16514682,007 ± 695689,124 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 16 thrpt 3 12196405,364 ± 882219,900 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 32 thrpt 3 8589420,490 ± 1266396,533 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 128 thrpt 3 3113779,210 ± 96236,236 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 512 thrpt 3 838036,695 ± 9111,954 ops/s EncodingHexJmhBenchmark.encodeOldBytesLib 1000000 thrpt 3 143,409 ± 77,980 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 4 thrpt 3 28925917,282 ± 1909118,263 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 8 thrpt 3 24633538,637 ± 1976249,711 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 16 thrpt 3 18627196,959 ± 161703,637 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 32 thrpt 3 14708733,929 ± 5488378,578 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 128 thrpt 3 6640536,630 ± 208149,440 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 512 thrpt 3 1865910,408 ± 36249,944 ops/s EncodingHexJmhBenchmark.encodeSchallinsky 1000000 thrpt 3 875,940 ± 608,061 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 4 thrpt 3 30130846,282 ± 3670664,087 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 8 thrpt 3 25747757,359 ± 1837807,768 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 16 thrpt 3 19435881,479 ± 4796200,589 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 32 thrpt 3 15339205,263 ± 481716,466 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 128 thrpt 3 6732995,825 ± 150642,614 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 512 thrpt 3 1891953,715 ± 26048,908 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode1 1000000 thrpt 3 887,777 ± 125,344 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 4 thrpt 3 31648591,182 ± 953394,103 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 8 thrpt 3 26786077,183 ± 225195,917 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 16 thrpt 3 20598995,852 ± 1062913,974 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 32 thrpt 3 15871946,268 ± 992030,494 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 128 thrpt 3 7375691,213 ± 2440196,295 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 512 thrpt 3 2147273,997 ± 32629,513 ops/s EncodingHexJmhBenchmark.encodeStackOverflowCode2 1000000 thrpt 3 1049,595 ± 93,412 ops/s
Process finished with exit code 0