Markk116/smarm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
2b85ef60b2a440edadf4a6e63867faa542042a18
smarm/benches/baseline-output/sweep/ai128_tc150k.txt
Bench 3da6ffaa77 benches: expose preemption knobs + sweep runner 
Config API changes (src/preempt.rs, src/runtime.rs):
- preempt: promote ALLOC_INTERVAL and TIMESLICE_CYCLES from bare consts to
  DEFAULT_ALLOC_INTERVAL / DEFAULT_TIMESLICE_CYCLES; store active values in
  thread-locals set on each actor resume so multiple runtimes can use
  different settings concurrently.
- runtime: add alloc_interval / timeslice_cycles fields to Config; add
  Config::alloc_interval(n) and Config::timeslice_cycles(c) builder methods;
  thread the values through RuntimeInner to the reset_timeslice() call in
  schedule_loop.

Bench changes:
- Add bench_cfg(threads) helper to general/tokio_favored/smarm_favored that
  wraps Config::exact and reads SMARM_ALLOC_INTERVAL / SMARM_TIMESLICE_CYCLES
  env vars, so the sweep script can vary knobs without recompiling.

Sweep tooling (benches/sweep.py):
- 'run':     run the 3-file bench suite once; --save-baseline persists JSON
- 'regress': compare current run against baseline.json, exit 1 on any bench
             that regresses >10% vs stored medians
- 'sweep':   run the full SWEEP_GRID (10 points), print comparison table,
             optional --save-csv; binaries pre-built so no recompile per point

Sweep results (10-point grid, 1-CPU sandbox):
- The preemption knobs have very little effect on this single-CPU machine.
  Most benches move <5% across the entire grid.
- Longer timeslices (tc=600k, tc=1200k) reliably hurt spawn_storm_busy
  (+11-15%) and catch_unwind_panics (+10-12%) because actors hold the
  scheduler mutex longer per timeslice, stalling the storm of joinable tasks.
- Shorter timeslices (tc=150k) give a small improvement on many_timers
  (-3-4%) and a wash everywhere else.
- yield_in_hot_loop and uncontended_channel are essentially flat across all
  knobs — both are scheduling-dominated and call yield_now explicitly, so
  the RDTSC-driven preemption path is irrelevant.
- Conclusion: the knobs matter primarily under contention (multi-core).
  Re-run sweep on a multi-core machine before drawing tuning conclusions.
2026-05-25 13:04:58 +00:00

127 lines
8.3 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8596 | 8491 | 8805
smarm 1-thread | 1000 | 8552 | 8461 | 9003
tokio current_thread | 1000 | 125 | 125 | 260
tokio multi-thread | 1000 | 190 | 184 | 338
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 41885 | 41112 | 43292
smarm 1-thread | 200000 | 42174 | 41063 | 43145
tokio current_thread | 200000 | 15195 | 15010 | 15589
tokio multi-thread | 200000 | 16037 | 15869 | 17057
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29872 | 29629 | 31596
smarm 1-thread | 33860 | 29776 | 29528 | 30003
tokio current_thread | 33860 | 28705 | 28605 | 30287
tokio multi-thread | 33860 | 34655 | 34503 | 36596
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16898 | 16574 | 17386
smarm 1-thread | 1000 | 16871 | 16677 | 18467
tokio current_thread | 1000 | 897 | 857 | 991
tokio multi-thread | 1000 | 4325 | 4228 | 4458
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 133462 | 129526 | 138685
smarm 1-thread | 10000 | 130118 | 127633 | 142344
tokio current_thread | 10000 | 2713 | 2608 | 2831
tokio multi-thread | 10000 | 7367 | 4345 | 11741
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9077 | 8944 | 9287
smarm 1-thread | 320000 | 9100 | 9033 | 10604
tokio current_thread | 320000 | 17310 | 17122 | 18616
tokio multi-thread | 320000 | 17484 | 17413 | 17748
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 140039 | 135577 | 145123
smarm 1-thread | 10000 | 139931 | 135513 | 143841
tokio current_thread | 10000 | 14524 | 14378 | 14564
tokio multi-thread | 10000 | 15066 | 14677 | 15336
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29620 | 29511 | 31347
tokio multi 1-thread | 33860 | 29046 | 28817 | 29687
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 94 | 79 | 371
smarm 1-thread | 1 | 183 | 83 | 317
tokio current_thread | 1 | 25 | 25 | 31
tokio multi-thread | 1 | 54 | 41 | 71
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 189034 | 187674 | 192204
tokio current_thread | 1000000 | 151106 | 149564 | 155601
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 26949 | 26838 | 30868
tokio current_thread | 1000000 | 52984 | 52149 | 55141
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 145860 | 143015 | 152734
smarm 1-thread | 10000 | 144550 | 141592 | 149247
tokio current_thread | 10000 | 267500 | 265301 | 278751
tokio multi-thread | 10000 | 275320 | 268986 | 286891
Reference in New Issue View Git Blame Copy Permalink