YJIT Benchmarks

Details for Benchmarks at 2022-11-04 20:10:38

Overall YJIT is 33.2% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 30.5% faster than CRuby!
The basic "faster" measurement is the geomean of all "headlining" x86 benchmarks on this page.

Headlining Benchmarks

These are "headlining" because the "overall" speedup above is based on these benchmarks specifically.

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

Other Benchmarks

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

MicroBenchmarks

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

Want Raw Graphs and CSV?

Benchmarks Speed Details

Benchmark Memory Usage Details

Number of Iterations and Warmups Tested

bench No JIT warmups No JIT iters YJIT warmups YJIT iters
activerecord 5 204 5 204
hexapdf 5 15 5 15
liquid-render 5 175 5 175
mail 5 125 5 125
psych-load 5 15 5 15
railsbench 5 15 5 15
ruby-lsp 5 247 5 247
binarytrees 5 83 5 83
chunky_png 5 32 5 32
erubi 5 68 5 68
erubi_rails 5 1082 5 1082
etanni 5 42 5 42
fannkuchredux 5 15 5 15
lee 5 20 5 20
nbody 5 243 5 243
optcarrot 5 15 5 15
rubykon 5 15 5 15
30k_ifelse 5 50 5 50
30k_methods 5 21 5 21
cfunc_itself 5 469 5 469
fib 5 327 5 327
getivar 5 426 5 426
keyword_args 5 386 5 386
respond_to 5 710 5 710
setivar 5 400 5 400
str_concat 5 481 5 481

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
activerecord (click) 2323169 2321103 23 374 6 1% 0 0 0
hexapdf (click) 1777812 1777319 818 10181 189 1% 0 0 0
liquid-render (click) 596091 593657 154 1791 34 1% 0 0 0
mail (click) 1333310 1330885 371 5121 176 3% 0 0 0
psych-load (click) 808954 807071 68 583 19 3% 0 0 0
railsbench (click) 5659678 5658750 1777 13867 483 3% 0 0 0
ruby-lsp (click) 5601987 5601020 6423 41480 1520 3% 12604 0 0
binarytrees (click) 130426 127888 14 89 4 4% 0 0 0
chunky_png (click) 817224 814962 87 1134 31 2% 0 0 0
erubi (click) 747616 746737 11 97 4 4% 0 0 0
erubi_rails (click) 4754660 4752429 327 2623 111 4% 0 0 0
etanni (click) 131305 131162 11 57 4 7% 0 0 0
fannkuchredux (click) 142459 141267 11 219 4 1% 0 0 0
lee (click) 833467 831447 77 759 25 3% 0 0 0
nbody (click) 137584 135729 13 198 6 3% 0 0 0
optcarrot (click) 564934 564902 201 4024 93 2% 0 0 0
rubykon (click) 280926 279909 143 1595 32 2% 0 0 0
30k_ifelse (click) 5414509 5413169 9266 57146 4 0% 0 0 0
30k_methods (click) 2082775 2081467 5785 19375 4 0% 0 0 0
cfunc_itself (click) 128330 126281 11 72 4 5% 0 0 0
fib (click) 126876 125394 11 61 4 6% 0 0 0
getivar (click) 128947 127370 11 88 4 4% 0 0 0
keyword_args (click) 129123 126792 12 74 4 5% 0 0 0
respond_to (click) 128850 126884 11 88 5 5% 0 0 0
setivar (click) 129114 126821 11 61 4 6% 0 0 0
str_concat (click) 128573 126924 13 87 6 6% 0 0 0

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.