gavindiaz/polyfill-rs
1
0
Fork 0
mirror of https://github.com/floor-licker/polyfill-rs.git synced 2026-08-16 14:08:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

perf: retain snapshots by generation

Browse Source
This commit is contained in:
floor-licker
2026-06-22 13:46:29 -04:00
parent ebc2929b1e
commit ddbacf5fbb
3 changed files with 102 additions and 104 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+1 -1
View File
@@ -71,7 +71,7 @@ Real-world Polymarket API latency broken down by request phase:
| **Order Book Updates (1000 ops)** | 159.6 µs ± 32 µs | 6,260 updates/sec, zero-allocation |
| **Spread/Mid Calculations** | 70 ns ± 77 ns | 14.3M ops/sec, optimized BTreeMap |
| **JSON Parsing (480KB)** | ~2.3 ms | SIMD-accelerated parsing (1.77x faster than serde_json) |
| **WS `book` hot path (decode + apply)** | ~0.27 µs / 7.46 µs / 93.24 µs | 1 / 16 / 64 levels-per-side, strict 4dp fixed-point tape parser, no Decimal/rounding/clamping in the feed path (see `benches/ws_hot_path.rs`) |
| **WS `book` hot path (decode + apply)** | ~0.23 µs / 1.73 µs / 6.74 µs | 1 / 16 / 64 levels-per-side, strict fixed-point tape parser with generation-marked snapshot retention (see `benches/ws_hot_path.rs`) |
Run the WS hot-path benchmark locally with `cargo bench --bench ws_hot_path`.