YJIT Benchmarks

Details for Benchmarks at 2022-04-03 07:13:34

Overall YJIT is 32.3% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 30.3% 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 MJIT warmups MJIT iters YJIT warmups YJIT iters
activerecord 5 179 75 179 20 179
hexapdf 5 15 20 15
liquid-render 5 154 75 154 20 154
mail 5 123 75 123 20 123
psych-load 5 15 75 15 20 15
railsbench 5 15 75 15 20 15
binarytrees 5 65 75 65 20 65
chunky_png 5 30 75 30 20 30
discourse 5 37 75 37 20 37
erubi 5 47 75 47 20 47
erubi_rails 5 977 75 977 20 977
fannkuchredux 5 15 72 15 20 15
lee 5 21 75 21 20 21
nbody 5 281 75 281 20 281
optcarrot 5 15 75 15 20 15
rubykon 5 15 22 15 20 15
30k_ifelse 5 56 75 56 20 56
30k_methods 5 22 29 22 20 22
cfunc_itself 5 443 75 443 20 443
fib 5 320 75 320 20 320
getivar 5 771 75 771 20 771
keyword_args 5 377 75 377 20 377
respond_to 5 103 75 103 20 103
setivar 5 1917 75 1917 20 1917

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) 993041 814710 185 1185 0 0% 0 0 0
hexapdf (click) 1099025 922604 599 7882 428 5% 0 0 321
liquid-render (click) 502865 425097 143 1517 92 6% 0 0 5
mail (click) 844625 656183 324 4961 96 1% 0 0 111
psych-load (click) 341713 286821 59 434 1 0% 0 0 0
railsbench (click) 2600081 2105813 1297 9400 224 2% 0 0 26
binarytrees (click) 170257 139155 11 58 0 0% 0 0 0
chunky_png (click) 376337 309840 85 867 0 0% 0 0 0
discourse (click) 6204113 5573358 3088 26520 3131 11% 2 0 140
erubi (click) 329233 276912 10 79 0 0% 0 0 0
erubi_rails (click) 1960081 1599618 262 1891 12 0% 0 0 3
fannkuchredux (click) 183953 150015 8 188 0 0% 0 0 0
lee (click) 378769 315736 45 582 68 11% 0 0 13
nbody (click) 179857 146734 10 158 0 0% 0 0 0
optcarrot (click) 507089 452364 198 3575 20 0% 0 0 0
rubykon (click) 311249 259111 140 1529 1 0% 0 0 0
30k_ifelse (click) 5558801 4359150 9263 57804 0 0% 0 0 0
30k_methods (click) 2174801 1668319 5782 19361 0 0% 0 0 0
cfunc_itself (click) 168465 137524 8 50 0 0% 0 0 0
fib (click) 167313 137392 8 38 0 0% 0 0 0
getivar (click) 169361 140208 8 65 0 0% 0 0 0
keyword_args (click) 169937 138803 9 51 0 0% 0 0 0
respond_to (click) 170833 139400 8 64 0 0% 0 0 0
setivar (click) 168977 138589 8 38 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.