Razvan Dimescu 50828c411a fix: cold benchmark uses 1 round per domain for genuine cold measurements ...

With ROUNDS=10, only the first query per domain was truly cold — the
other 9 hit cached NS delegations at <1ms, diluting the median to
0.4ms. Now cold mode uses 1 round so every sample is a real cold
resolve. Also extracted compare_two_rounds to support per-mode rounds.

2026-04-12 21:00:24 +03:00

..

dnssec.rs

feat: recursive DNS + DNSSEC + TCP fallback (#17)

2026-03-28 04:03:47 +02:00

hot_path.rs

perf: optimize DNS query hot path (#15)

2026-03-27 02:01:08 +02:00

numa-bench-recursive.toml

bench: add --vs-nextdns, --vs-unbound-cold modes with mode validation

2026-04-12 18:41:09 +03:00

numa-bench.toml

feat: DoT client, recursive optimization, bench refactor

2026-04-12 18:40:46 +03:00

recursive_compare.rs

fix: cold benchmark uses 1 round per domain for genuine cold measurements

2026-04-12 21:00:24 +03:00

throughput.rs

perf: optimize DNS query hot path (#15)

2026-03-27 02:01:08 +02:00