YJIT Benchmarks

Details for Benchmarks at 2021-09-28 07:09:45

Overall YJIT is 23.5% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 20.4% 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 10 125 10 119 10 166
liquid-render 10 98 10 109 10 126
mail 10 120 10 122 10 134
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 15
nbody 10 169 10 331 10 265
optcarrot 10 10 10 10 10 10
30k_ifelse 10 10 10 10 10 50
30k_methods 10 10 10 10 10 17
cfunc_itself 10 215 10 339 10 473
fib 10 91 10 303 10 375
getivar 10 218 10 210 10 529
respond_to 10 74 10 88 10 110
setivar 10 289 10 265 10 469

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) 1085082 833334 185 1148 0 0% 0 0 0
liquid-render (click) 437844 352614 143 1211 92 7% 0 0 1
mail (click) 969498 743717 337 5141 129 2% 0 0 39
psych-load (click) 175082 138677 56 408 1 0% 0 0 0
railsbench (click) 2498353 1917129 1293 9346 255 2% 0 0 26
binarytrees (click) 106429 82739 9 49 0 0% 0 0 0
fannkuchredux (click) 122493 96364 6 203 0 0% 0 0 0
jekyll (click) 1935706 1590970 336 3751 173 4% 0 0 0
lee (click) 223722 169830 85 1046 111 10% 1 0 13
nbody (click) 116477 90863 8 149 0 0% 0 0 0
optcarrot (click) 642813 576070 286 5570 20 0% 0 0 0
30k_ifelse (click) 5554237 4883816 9261 57795 0 0% 0 0 0
30k_methods (click) 2110589 1809485 5780 19352 0 0% 0 0 0
cfunc_itself (click) 104957 81178 6 40 0 0% 0 0 0
fib (click) 103357 80751 6 29 0 0% 0 0 0
getivar (click) 105149 83592 6 56 0 0% 0 0 0
respond_to (click) 107133 82722 6 55 0 0% 0 0 0
setivar (click) 105341 82049 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.