← Back to comparison

Ferrous DNS

Port: 15353 | Image: ferrous:bench

Results by Scenario

Cache hit

All queries resolve from server cache — measures peak throughput.

Provider QPS Avg (ms) Lost % Total Qrys Base Mem (MB)
Quad9 957.81 96.91 0.14 28858 143.36
Cloudflare 912.15 96.93 0.25 27522 143.52
Google 921.44 92.47 0.38 27837 116.48
OpenDNS 958.06 97.25 0.13 28935 136.61
Quad9 958.79 95.79 0.17 28902 141.18
Cloudflare 1005.41 94.23 0.10 30287 142.04
Google 996.56 93.96 0.11 30015 141.59
OpenDNS 999 96.05 0.09 30087 142.66

Cache miss

Every query requires full upstream resolution — measures worst-case performance. In common-denominator mode some servers report near-zero QPS when plain-DNS upstream forwarding fails.

Provider QPS Avg (ms) Lost % Total Qrys Base Mem (MB)
Quad9 198.74 104.96 0.06 6000 143.26
Cloudflare 199.56 103.56 0.04 6000 143.34
Google 197.59 99.18 0.35 5959 132.33
OpenDNS 199.54 102.21 0.06 6000 141.09
Quad9 175.30 208.32 0.32 5285 139.82
Cloudflare 199.23 106.64 0.19 6000 141.65
Google 199.48 101.79 0.07 6000 140.99
OpenDNS 199.60 103.05 0.01 6000 141.72

Mixed

90% cached + 10% unique domains — simulates real-world mixed workload.

Provider QPS Avg (ms) Lost % Total Qrys Base Mem (MB)
Quad9 935.90 98.70 0.16 28218 146.36
Cloudflare 900.28 98.64 0.25 27155 146.38
Google 935.71 96.13 0.21 28216 135.51
OpenDNS 983.51 96.83 0.08 29620 145.60
Quad9 943.58 95.53 0.23 28460 143.71
Cloudflare 1014.45 93.69 0.09 30555 144.18
Google 986.07 97.32 0.08 29695 144.38
OpenDNS 993.57 96.16 0.08 29922 145.10

Blocked

Domains on the sinkhole blocklist — measures filtering throughput.

Provider QPS Avg (ms) Lost % Total Qrys Base Mem (MB)
Quad9 956.52 99.76 0.10 28808 150.54
Cloudflare 967.51 94.95 0.20 29185 148.98
Google 915.23 97.03 0.27 27602 138.62
OpenDNS 981.16 94.60 0.15 29567 149.10
Quad9 898.89 100.70 0.22 27104 148.09
Cloudflare 985.50 94.68 0.14 29694 148.11
Google 997.28 93.08 0.14 30059 147.91
OpenDNS 1003.27 96.05 0.12 30222 147.69

Methodology

Transport Protocol Confound

This server uses DoH for best-of-breed mode. Comparisons against servers using different protocols (DoH, DoT, plain) partly reflect protocol overhead. The common-denominator mode forces all servers to plain UDP 53 to isolate server-side performance.

Warmup

Before each scenario, a 5-second dnsperf warmup pass is run to stabilize connections and fill the TCP congestion window. For cache-hit and mixed scenarios, the server cache is also pre-warmed. The measurement window (-l 30) begins only after both warmup phases complete.

Test Parameters

Back to comparison | Generated 2026-07-01T21:33:06Z