YJIT Benchmarks

Details for Benchmarks at 2022-10-18 04:42:22

Overall YJIT is 37.0% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 34.3% 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 YJIT warmups YJIT iters
activerecord 5 203 5 203
hexapdf 5 15 5 15
liquid-render 5 172 5 172
mail 5 128 5 128
psych-load 5 15 5 15
railsbench 5 15 5 15
binarytrees 5 86 5 86
chunky_png 5 31 5 31
erubi 5 67 5 67
erubi_rails 5 1051 5 1051
etanni 5 42 5 42
fannkuchredux 5 15 5 15
lee 5 20 5 20
nbody 5 244 5 244
optcarrot 5 15 5 15
rubykon 5 15 5 15
30k_ifelse 5 57 5 57
30k_methods 5 24 5 24
cfunc_itself 5 469 5 469
fib 5 325 5 325
getivar 5 426 5 426
keyword_args 5 386 5 386
respond_to 5 707 5 707
setivar 5 424 5 424
str_concat 5 408 5 408

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) 4642930 4648945 23 319 6 1% 0 0 0
hexapdf (click) 3671239 3678953 817 9679 189 1% 0 0 0
liquid-render (click) 1429066 1435891 154 1742 34 1% 0 0 0
mail (click) 2835695 2843369 371 4957 176 3% 0 0 0
psych-load (click) 1777018 1783270 68 489 19 3% 0 0 0
railsbench (click) 10226138 10233400 1777 12850 475 3% 0 0 0
binarytrees (click) 590037 598090 14 70 4 5% 0 0 0
chunky_png (click) 1825030 1831352 87 1034 31 2% 0 0 0
erubi (click) 1689505 1697144 11 78 4 5% 0 0 0
erubi_rails (click) 8440344 8447544 327 2448 107 4% 0 0 0
etanni (click) 591909 599381 11 38 4 10% 0 0 0
fannkuchredux (click) 611538 617988 11 200 4 2% 0 0 0
lee (click) 1852346 1860115 77 746 25 3% 0 0 0
nbody (click) 612932 619097 13 179 6 3% 0 0 0
optcarrot (click) 1459687 1467449 201 4001 93 2% 0 0 0
rubykon (click) 888131 895184 143 1564 32 2% 0 0 0
30k_ifelse (click) 11178938 11185449 9266 57127 4 0% 0 0 0
30k_methods (click) 4510782 4517361 5785 19356 4 0% 0 0 0
cfunc_itself (click) 580808 585658 11 53 4 7% 0 0 0
fib (click) 577901 584541 11 42 4 9% 0 0 0
getivar (click) 579980 586390 11 69 4 5% 0 0 0
keyword_args (click) 580150 585499 12 55 4 7% 0 0 0
respond_to (click) 578702 585293 11 55 5 9% 0 0 0
setivar (click) 581593 586483 11 42 4 9% 0 0 0
str_concat (click) 579567 585755 13 68 6 8% 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.