YJIT Benchmarks

Details for Benchmarks at 2024-01-10 06:08:17 UTC

YJIT metrics from the yjit-bench suite using Ruby e59a730477

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 64.0% faster than CRuby 3.4.0dev
On railsbench it is
  • 68.1% faster than CRuby 3.4.0dev

Performance on Headline Benchmarks

Speed of each Ruby implementation relative to the baseline CRuby measurement. Higher is better.

Memory Usage on Headline Benchmarks

Memory usage of each Ruby implementation relative to the baseline CRuby measurement. Lower is better.

Performance on Other Benchmarks

Speed of each Ruby implementation relative to the baseline CRuby measurement. Higher is better.

Memory Usage on Other Benchmarks

Memory usage of each Ruby implementation relative to the baseline CRuby measurement. Lower is better.

Performance on MicroBenchmarks

Speed of each Ruby implementation relative to the baseline CRuby measurement. Higher is better.

Memory Usage on MicroBenchmarks

Memory usage of each Ruby implementation relative to the baseline CRuby measurement. Lower is better.

Want Raw Graphs and CSV?

Benchmarks Speed Details

Benchmark Memory Usage Details

Number of Iterations and Warmups Tested

bench CRuby 3.4.0dev warmups CRuby 3.4.0dev iters YJIT 3.4.0dev warmups YJIT 3.4.0dev iters
activerecord 30 472 30 472
chunky-png 30 30 30 30
erubi-rails 30 15 30 15
hexapdf 30 15 30 15
liquid-c 30 344 30 344
liquid-compile 30 337 30 337
liquid-render 30 242 30 242
lobsters 30 23 30 23
mail 30 164 30 164
psych-load 30 15 30 15
railsbench 30 15 30 15
ruby-lsp 30 388 30 388
sequel 30 293 30 293
binarytrees 30 78 30 78
erubi 30 75 30 75
etanni 30 50 30 50
fannkuchredux 30 24 30 24
fluentd 30 15 30 15
graphql 30 15 30 15
graphql-native 30 33 30 33
lee 30 19 30 19
matmul 30 15 30 15
nbody 30 278 30 278
nqueens 30 105 30 105
optcarrot 30 15 30 15
rack 30 176 30 176
ruby-json 30 15 30 15
rubykon 30 15 30 15
sudoku 30 15 30 15
tinygql 30 40 30 40
30k_ifelse 30 55 30 55
30k_methods 30 23 30 23
cfunc_itself 30 751 30 751
fib 30 416 30 416
getivar 30 940 30 940
keyword_args 30 493 30 493
respond_to 30 1184 30 1184
setivar 30 1668 30 1668
setivar_object 30 360 30 360
setivar_young 30 389 30 389
str_concat 30 400 30 400
throw 30 831 30 831

Different Ruby configurations want different amounts of warmup. With no JIT, CRuby needs hardly any. YJIT and MJIT 3.0 both warm up quite quickly, while MJIT in 3.1 often slows down for a time as it compiles, after an unpredictable delay.

Benchmark YJIT Stats

Hover your cursor over the benchmark names for descriptions of each benchmark.

bench Exit Report Inline Outlined Comp iSeqs Comp Blocks Inval Inval Ratio Bind Alloc Bind Set Const Bumps Compile Time MS
activerecord (click) 734933 642524 50 504 3 0% 0 0 0 14.690482
chunky-png (click) 307325 261267 87 1104 1 0% 0 0 0 33.452619
erubi-rails (click) 1611688 1400530 287 2903 1 0% 0 0 0 82.333635
hexapdf (click) 1330739 1141260 598 12085 36 0% 0 0 0 354.764976
liquid-c (click) 486269 444504 120 1701 5 0% 0 0 0 46.970402
liquid-compile (click) 348168 259280 150 2022 2 0% 0 0 0 61.669543
liquid-render (click) 519391 379452 143 2161 5 0% 0 0 0 60.230986
lobsters (click) 7400800 6467268 3182 48189 64 0% 0 0 0 1610.936422
mail (click) 682842 620025 345 4760 12 0% 0 0 0 140.232998
psych-load (click) 260964 220020 65 600 3 0% 0 0 0 18.7668
railsbench (click) 2458933 2160243 1357 12408 16 0% 0 0 0 362.957014
ruby-lsp (click) 5298378 4654760 3395 49689 397 0% 4 0 0 1520.658075
sequel (click) 454661 406389 16 127 0 0% 0 0 0 4.978585
binarytrees (click) 9306 8120 11 75 0 0% 0 0 0 2.932407
erubi (click) 232785 229083 10 95 0 0% 0 0 0 3.218859
etanni (click) 32063 22322 12 90 0 0% 0 0 0 3.048754
fannkuchredux (click) 22558 23293 9 233 0 0% 0 0 0 7.340212
fluentd (click) 332166 282602 12 104 0 0% 0 0 0 3.361383
graphql (click) 397052 350921 70 658 0 0% 0 0 0 20.144057
graphql-native (click) 362855 319578 42 264 0 0% 0 0 0 8.738896
lee (click) 281180 250100 49 691 0 0% 0 0 0 22.187791
matmul (click) 14070 12895 17 122 0 0% 0 0 0 4.418941
nbody (click) 18283 17348 11 189 0 0% 0 0 0 5.663823
nqueens (click) 18613 21105 12 196 0 0% 0 0 0 6.183688
optcarrot (click) 314630 260302 194 4242 32 0% 0 0 0 100.782086
rack (click) 239182 206297 31 304 0 0% 0 0 0 8.830663
ruby-json (click) 27788 25201 12 203 0 0% 0 0 0 6.240033
rubykon (click) 129011 118454 144 1473 3 0% 0 0 0 41.542499
sudoku (click) 44938 40709 26 421 0 0% 0 0 0 13.109334
tinygql (click) 277335 244871 63 773 5 0% 0 0 0 22.354328
30k_ifelse (click) 4909644 4374870 9264 50817 0 0% 0 0 0 1381.295921
30k_methods (click) 1850351 1594387 5782 19371 0 0% 0 0 0 493.038263
cfunc_itself (click) 7868 6931 9 73 0 0% 0 0 0 2.375695
fib (click) 5044 5012 8 49 0 0% 0 0 0 1.785315
getivar (click) 6179 6402 8 74 0 0% 0 0 0 2.107434
keyword_args (click) 8481 7140 10 75 0 0% 0 0 0 2.566309
respond_to (click) 8611 8551 9 88 0 0% 0 0 0 2.813073
setivar (click) 5330 4983 8 54 0 0% 0 0 0 1.809146
setivar_object (click) 5668 5003 8 54 0 0% 0 0 0 1.920031
setivar_young (click) 6435 5792 9 62 0 0% 0 0 0 2.094141
str_concat (click) 8079 7814 11 87 0 0% 0 0 0 2.818907
throw (click) 7723 6444 10 69 0 0% 0 0 0 2.396896

YJIT stats correspond to the YJIT stats exit report.

Note: currently, all stats are collected on x86_64, not ARM.

Raw JSON data files

All graphs and table data in this page comes from processing these data files, which come from benchmark runs.