YJIT Benchmarks

Details for Benchmarks at 2021-11-05 19:09:51

Overall YJIT is 27.3% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 19.0% 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 MJIT3.0 warmups MJIT3.0 iters YJIT warmups YJIT iters
activerecord 10 128 10 114 10 173
hexapdf 10 10 10 10 10 10
liquid-render 10 103 10 109 10 160
mail 10 114 10 124 10 126
psych-load 10 10 10 10 10 10
railsbench 10 10 10 10 10 10
binarytrees 10 48 10 70 10 62
fannkuchredux 10 10 10 10 10 10
jekyll 10 10 10 10 10 10
lee 10 10 10 14 10 15
nbody 10 178 10 331 10 270
optcarrot 10 10 10 10 10 10
rubykon 10 10 10 10 10 10
30k_ifelse 10 10 10 10 10 52
30k_methods 10 10 10 10 10 17
cfunc_itself 10 211 10 339 10 506
fib 10 94 10 307 10 375
getivar 10 228 10 211 10 531
keyword_args 10 74 10 97 10 422
respond_to 10 78 10 90 10 118
setivar 10 298 10 264 10 443

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) 978612 768823 184 1157 0 0% 0 0 0
hexapdf (click) 1095220 916656 603 7967 624 7% 0 0 320
liquid-render (click) 476353 389600 143 1428 96 6% 0 0 1
mail (click) 868404 673525 339 5257 124 2% 0 0 39
psych-load (click) 301665 241615 57 416 1 0% 0 0 0
railsbench (click) 2596849 2027900 1295 9512 262 2% 0 0 26
binarytrees (click) 114986 91464 9 48 0 0% 0 0 0
fannkuchredux (click) 131050 105089 6 202 0 0% 0 0 0
jekyll (click) 1902196 1581186 336 3884 177 4% 0 0 0
lee (click) 349492 273796 44 566 70 12% 0 0 12
nbody (click) 123818 98800 8 148 0 0% 0 0 0
optcarrot (click) 460906 431364 196 3566 20 0% 0 0 0
rubykon (click) 257962 223126 138 1518 1 0% 0 0 0
30k_ifelse (click) 5562538 4892393 9261 57794 0 0% 0 0 0
30k_methods (click) 2119146 1818210 5780 19351 0 0% 0 0 0
cfunc_itself (click) 112874 89509 6 39 0 0% 0 0 0
fib (click) 111914 89476 6 28 0 0% 0 0 0
getivar (click) 113706 92317 6 55 0 0% 0 0 0
keyword_args (click) 114282 90659 7 41 0 0% 0 0 0
respond_to (click) 115114 91053 6 54 0 0% 0 0 0
setivar (click) 113706 90626 6 28 0 0% 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.