YJIT Benchmarks

Details for Benchmarks at 2021-09-25 19:09:26

Overall YJIT is 22.4% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 21.3% 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 10 116 10 166
liquid-render 10 95 10 112 10 127
mail 10 121 10 123 10 129
psych-load 10 10 10 10 10 10
railsbench 10 10 10 10 10 10
binarytrees 10 42 10 69 10 57
fannkuchredux 10 10 10 10 10 10
jekyll 10 10 10 10 10 10
lee 10 10 10 13 10 16
nbody 10 170 10 331 10 265
optcarrot 10 10 10 10 10 10
30k_ifelse 10 10 10 10 10 53
30k_methods 10 10 10 10 10 17
cfunc_itself 10 208 10 329 10 471
fib 10 92 10 303 10 375
getivar 10 218 10 211 10 528
respond_to 10 72 10 90 10 108
setivar 10 291 10 265 10 470

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) 1090330 835343 185 1153 0 0% 0 0 0
liquid-render (click) 443988 355859 143 1251 92 7% 0 0 1
mail (click) 954778 729218 337 4938 125 2% 0 0 39
psych-load (click) 176938 139263 56 408 1 0% 0 0 0
railsbench (click) 2509489 1919182 1293 9353 256 2% 0 0 26
binarytrees (click) 107965 83079 9 49 0 0% 0 0 0
fannkuchredux (click) 123965 96704 6 203 0 0% 0 0 0
jekyll (click) 1948506 1593562 336 3749 173 4% 0 0 0
lee (click) 225258 170190 85 1045 111 10% 1 0 13
nbody (click) 117309 90809 8 149 0 0% 0 0 0
optcarrot (click) 644285 576410 286 5570 20 0% 0 0 0
30k_ifelse (click) 5555517 4884008 9261 57795 0 0% 0 0 0
30k_methods (click) 2112637 1810219 5780 19352 0 0% 0 0 0
cfunc_itself (click) 106365 81518 6 40 0 0% 0 0 0
fib (click) 104829 81091 6 29 0 0% 0 0 0
getivar (click) 106621 83932 6 56 0 0% 0 0 0
respond_to (click) 108605 83062 6 55 0 0% 0 0 0
setivar (click) 106621 82241 6 29 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.