YJIT Benchmarks

Details for Benchmarks at 2021-12-19 19:10:23

Overall YJIT is 32.0% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 31.1% 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 MJIT warmups MJIT iters YJIT warmups YJIT iters
activerecord 5 191 20 191 75 191 20 191
hexapdf 5 15 20 15 20 15
liquid-render 5 174 20 174 75 174 20 174
mail 5 143 20 143 75 143 20 143
psych-load 5 15 20 15 75 15 20 15
railsbench 5 15 20 15 75 15 20 15
binarytrees 5 86 20 86 75 86 20 86
erubi 5 53 20 53 75 53 20 53
erubi_rails 5 631 20 631 75 631 20 631
fannkuchredux 5 15 20 15 53 15 20 15
jekyll 5 15 20 15 36 15 20 15
lee 5 26 20 26 75 26 20 26
nbody 5 343 20 343 75 343 20 343
optcarrot 5 15 20 15 75 15 20 15
rubykon 5 15 20 15 44 15 20 15
30k_ifelse 5 61 20 61 75 61 20 61
30k_methods 5 27 20 27 45 27 20 27
cfunc_itself 5 491 20 491 75 491 20 491
fib 5 383 20 383 75 383 20 383
getivar 5 564 20 564 75 564 20 564
keyword_args 5 432 20 432 75 432 20 432
respond_to 5 125 20 125 75 125 20 125
setivar 5 505 20 505 75 505 20 505

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) 973674 741777 185 1194 0 0% 0 0 0
hexapdf (click) 1087594 886794 599 8110 618 7% 0 0 321
liquid-render (click) 467127 369168 143 1419 95 6% 0 0 1
mail (click) 856042 636528 325 5234 124 2% 0 0 39
psych-load (click) 292769 229848 57 423 1 0% 0 0 0
railsbench (click) 2506530 1894398 1295 9263 266 2% 0 0 26
binarytrees (click) 113381 86737 9 48 0 0% 0 0 0
erubi (click) 272289 213298 8 35 0 0% 0 0 0
erubi_rails (click) 2435682 1763635 261 1983 16 0% 0 0 3
fannkuchredux (click) 128997 98531 6 202 0 0% 0 0 0
jekyll (click) 1862058 1497056 339 3781 185 4% 0 0 0
lee (click) 330273 259277 43 557 66 11% 0 0 12
nbody (click) 122405 93722 8 148 0 0% 0 0 0
optcarrot (click) 451621 399534 196 3566 20 0% 0 0 0
rubykon (click) 255077 207015 138 1526 1 0% 0 0 0
30k_ifelse (click) 5502501 4307061 9261 57794 0 0% 0 0 0
30k_methods (click) 2117093 1615194 5780 19351 0 0% 0 0 0
cfunc_itself (click) 111397 85037 6 39 0 0% 0 0 0
fib (click) 109861 84645 6 28 0 0% 0 0 0
getivar (click) 111589 87156 6 55 0 0% 0 0 0
keyword_args (click) 112805 86007 7 41 0 0% 0 0 0
respond_to (click) 114085 86820 6 54 0 0% 0 0 0
setivar (click) 111717 85793 6 28 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.