YJIT Benchmarks

Details for Benchmarks at 2021-09-19 19:09:10

Overall YJIT is 24.4% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 18.4% 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 10 121 10 119 10 172
liquid-render 10 96 10 111 10 128
mail 10 120 10 121 10 127
psych-load 10 10 10 10 10 10
railsbench 10 10 10 10 10 10
binarytrees 10 43 10 69 10 58
fannkuchredux 10 10 10 10 10 10
jekyll 10 10 10 10 10 10
lee 10 10 10 13 10 15
nbody 10 170 10 331 10 266
optcarrot 10 10 10 10 10 10
30k_ifelse 10 10 10 10 10 49
30k_methods 10 10 10 10 10 19
cfunc_itself 10 211 10 338 10 478
fib 10 85 10 308 10 340
getivar 10 201 10 211 10 527
respond_to 10 76 10 90 10 113
setivar 10 291 10 260 10 438

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) 1076952 821738 185 1143 0 0% 0 0 0
liquid-render (click) 438482 352156 143 1225 92 7% 0 0 1
mail (click) 905816 693113 337 4550 125 2% 0 0 39
psych-load (click) 173863 137063 56 398 1 0% 0 0 0
railsbench (click) 5183340 3076293 1292 49137 40248 81% 0 0 26
binarytrees (click) 107450 82474 9 49 0 0% 0 0 0
fannkuchredux (click) 123002 95758 6 203 0 0% 0 0 0
jekyll (click) 2347672 1747556 335 9948 6515 65% 0 0 0
lee (click) 223975 169479 85 1045 111 10% 1 0 13
nbody (click) 116346 89863 8 149 0 0% 0 0 0
optcarrot (click) 641338 575411 286 5570 20 0% 0 0 0
30k_ifelse (click) 5527610 4883403 9261 57795 0 0% 0 0 0
30k_methods (click) 2099578 1809220 5780 19352 0 0% 0 0 0
cfunc_itself (click) 105914 80913 6 40 0 0% 0 0 0
fib (click) 104058 80240 6 29 0 0% 0 0 0
getivar (click) 106234 83327 6 56 0 0% 0 0 0
respond_to (click) 107834 82211 6 55 0 0% 0 0 0
setivar (click) 105850 81390 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.