YJIT Benchmarks

Details for Benchmarks at 2021-09-23 19:09:32

Overall YJIT is 25.0% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 18.7% 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 10 120 10 120 10 166
liquid-render 10 98 10 112 10 127
mail 10 117 10 125 10 134
psych-load 10 10 10 10 10 10
railsbench 10 10 10 10 10 10
binarytrees 10 40 10 68 10 58
fannkuchredux 10 10 10 10 10 10
jekyll 10 10 10 10 10 10
lee 10 10 10 13 10 15
nbody 10 167 10 332 10 271
optcarrot 10 10 10 10 10 10
30k_ifelse 10 10 10 10 10 53
30k_methods 10 10 10 10 10 17
cfunc_itself 10 211 10 335 10 466
fib 10 82 10 308 10 375
getivar 10 205 10 211 10 528
respond_to 10 78 10 88 10 117
setivar 10 282 10 265 10 443

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) 1087450 826880 185 1154 0 0% 0 0 0
liquid-render (click) 440276 351747 143 1221 92 7% 0 0 1
mail (click) 953370 725624 337 4938 125 2% 0 0 39
psych-load (click) 176362 137965 56 408 1 0% 0 0 0
railsbench (click) 2502897 1900978 1293 9342 255 2% 0 0 26
binarytrees (click) 108285 82474 9 49 0 0% 0 0 0
fannkuchredux (click) 123965 95853 6 203 0 0% 0 0 0
jekyll (click) 1948890 1581361 336 3769 173 4% 0 0 0
lee (click) 225706 169479 85 1045 111 10% 1 0 13
nbody (click) 117693 90204 8 149 0 0% 0 0 0
optcarrot (click) 643965 575316 286 5570 20 0% 0 0 0
30k_ifelse (click) 5555965 4883403 9261 57795 0 0% 0 0 0
30k_methods (click) 2111933 1808879 5780 19352 0 0% 0 0 0
cfunc_itself (click) 106237 80572 6 40 0 0% 0 0 0
fib (click) 104701 80145 6 29 0 0% 0 0 0
getivar (click) 106621 83081 6 56 0 0% 0 0 0
respond_to (click) 108989 82457 6 55 0 0% 0 0 0
setivar (click) 106493 81295 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.