0 Fork 0

mirror of https://github.com/PR0M3TH3AN/TORCH.git synced 2026-11-10 06:42:21 +00:00

Files

T

google-labs-jules[bot]andPR0M3TH3AN 6614535361 perf: optimize rankRelays for subset queries

- Optimized `RelayHealthManager.rankRelays` to use O(M log M) lookup for small subsets instead of O(N) filter.
- Added `byRelay` Map to `_sortedCache` for O(1) access.
- Fixed `NaN` bug in `buildRelayHealthConfig` when `minActiveRelayPool` is undefined.
- Updated `perf/benchmark-lock-ops.mjs` to measure subset performance.
- Added performance report `reports/performance/weekly-perf-report-2026-02-18.md`.
- Achieved ~6.8x speedup in benchmark.

Co-authored-by: PR0M3TH3AN <[email protected]>

2026-02-18 19:54:47 +00:00

470 B

Raw Blame History

Weekly Performance Report: 2026-02-18

System Info

Linux devbox 6.8.0 #1 SMP PREEMPT_DYNAMIC Wed Jan 14 23:11:35 UTC 2026 x86_64 x86_64 x86_64 GNU/Linux v22.22.0

Baseline: Lock Ops

RankRelays x 10000: 209.17ms
Average per call: 0.0209ms

Baseline: Lock Ops (Subset 5/1000)

RankRelays (Subset 5/1000) x 10000: 237.77ms
Average per call: 0.0238ms

After: Lock Ops

RankRelays (Subset 5/1000) x 10000: 34.78ms
Average per call: 0.0035ms

470 B Raw Blame History