YJIT Benchmarks

Details for Benchmarks at 2021-10-02 19:09:12

Overall YJIT is 23.2% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 21.1% 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 10 129 10 120 10 173
liquid-render 10 104 10 112 10 126
mail 10 123 10 123 10 136
psych-load 10 10 10 10 10 10
railsbench 10 10 10 10 10 10
binarytrees 10 43 10 67 10 57
fannkuchredux 10 10 10 10 10 10
jekyll 10 10 10 10 10 10
lee 10 10 10 13 10 15
nbody 10 172 10 325 10 267
optcarrot 10 10 10 10 10 10
30k_ifelse 10 10 10 10 10 52
30k_methods 10 10 10 10 10 17
cfunc_itself 10 217 10 326 10 492
fib 10 89 10 307 10 375
getivar 10 199 10 211 10 528
keyword_args 10 70 10 97 10 73
respond_to 10 75 10 90 10 105
setivar 10 280 10 264 10 492

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) 1081626 836669 185 1143 0 0% 0 0 0
liquid-render (click) 435924 354149 143 1206 92 7% 0 0 1
mail (click) 951066 731932 337 4962 129 2% 0 0 39
psych-load (click) 175146 139378 56 408 1 0% 0 0 0
railsbench (click) 2496753 1930431 1293 9378 255 2% 0 0 26
binarytrees (click) 106685 83176 9 49 0 0% 0 0 0
fannkuchredux (click) 122749 96801 6 203 0 0% 0 0 0
jekyll (click) 1933914 1605921 336 3730 174 4% 0 0 0
lee (click) 223402 169914 85 1040 111 10% 1 0 13
nbody (click) 116477 91152 8 149 0 0% 0 0 0
optcarrot (click) 643261 577913 286 5575 20 0% 0 0 0
30k_ifelse (click) 5554109 4883957 9261 57795 0 0% 0 0 0
30k_methods (click) 2110781 1809922 5780 19352 0 0% 0 0 0
cfunc_itself (click) 105149 81615 6 40 0 0% 0 0 0
fib (click) 103997 81434 6 29 0 0% 0 0 0
getivar (click) 105213 83881 6 56 0 0% 0 0 0
keyword_args (click) 102781 80220 7 22 0 0% 0 0 0
respond_to (click) 107197 83011 6 55 0 0% 0 0 0
setivar (click) 105789 82584 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.