RunsOn m8azn.xlarge · gp3 1000 MiB/s vs Namespace 4x16 x64
RunsOn does each build for 1.5–1.9× less; neither is clearly faster (within 5% or inside each other's run-to-run range). Both 4 vCPU x64.
All three builds 1.6× cheaper on RunsOn: $0.0568 vs $0.0900Namespace 1.4× faster: 7m 02s vs 9m 33s
Rust build 1.9× cheaper on RunsOn: $0.0104 vs $0.0198within run-to-run spread: 1m 41s · 1m 33s
TypeScript build 1.8× cheaper on RunsOn: $0.0060 vs $0.0106within run-to-run spread: 47s · 45s
Docker build 1.5× cheaper on RunsOn: $0.0404 vs $0.0596within run-to-run spread: 7m 05s · 4m 44s
RunsOn m8azn.xlarge · gp3 1000 MiB/s
jobs ran from Amazon.com (US)
Namespace 4x16
jobs ran from Namespace Labs (US)
Side by side
Every measurement both runners share, cost first. The dot marks the better value; gap is the larger over the smaller.
| Measurement | RunsOn | Namespace | Gap |
|---|---|---|---|
| Cost | |||
| Cost per Rust build | $0.0104 | $0.0198 | 1.9× RunsOn |
| Cost per TypeScript build | $0.0060 | $0.0106 | 1.8× RunsOn |
| Cost per Docker build | $0.0404 | $0.0596 | 1.5× RunsOn |
| Price per minute | $0.0045 | $0.0120 overage rate | 2.7× RunsOn |
| Real builds, wall time | |||
| Rust build (spotify-player) | 1m 41s | 1m 33s | ≈ even |
| TypeScript build (opencode) | 47s | 45s | ≈ even |
| Docker build (PostHog) | 7m 05s | 4m 44s | ≈ even |
| CPU & memory | |||
| PassMark single-thread (points) | 4,260 | 4,628 | 1.1× Namespace |
| PassMark CPU Mark, all cores (points) | 15,109 | 15,965 | 1.1× Namespace |
| sysbench CPU, 1 thread (events/s) | 5,619 | 6,887 | 1.2× Namespace |
| 7-Zip compression (MIPS) | 42,321 | 33,572 | 1.3× RunsOn |
| Memory read bandwidth | 41.0 GiB/s | 44.7 GiB/s | 1.1× Namespace |
| Measurement | RunsOn | Namespace | Gap |
|---|---|---|---|
| Workspace disk (fio) | |||
| Random read, 4k QD32 (IOPS) | 15,994 | 1,202,685 | 75.2× Namespace |
| Random write, 4k QD32 (IOPS) | 15,991 | 564,373 | 35.3× Namespace |
| Random read, 4k QD1 (IOPS) | 1,772 | 104,443 | 58.9× Namespace |
| QD1 read latency, p99 | 799 µs | 11 µs | 74.0× Namespace |
| Sequential read | 1,001 MiB/s | 5.0 GiB/s | 5.1× Namespace |
| Sequential write | 1,001 MiB/s | 3.4 GiB/s | 3.5× Namespace |
| Sync write, 4k + fdatasync (IOPS) | 541 | 28,561 | 52.8× Namespace |
| Network, cache & Docker | |||
| Download from GitHub | 172 MiB/s | 187 MiB/s | 1.1× Namespace |
| Cache restore, 4 GB (actions/cache) | 7s | 21s | 2.9× RunsOn |
| Cache save, 4 GB (actions/cache) | 9s | 13s | 1.4× RunsOn |
| docker pull | 4.6s | 3.2s | 1.4× Namespace |
| Database | |||
| pgbench read/write (TPS) | 3,741 | 13,662 | 3.7× Namespace |
| pgbench read-only (TPS) | 164,328 | 114,631 | 1.4× RunsOn |
| Queue | |||
| Queue under burst load | 17s | 10s | 1.6× Namespace |
- ≈ even: under 5% apart, or a build inside both runners' run-to-run spread
- Queue: identical jobs queued at the same instant (burst suite)
- RunsOn at EC2 spot, its default (7-day average of the cheapest us-east-1 zone's spot price, fetched 16 Nov 2026); its jobs ran on spot
- Cost: a managed SaaS runner against RunsOn self-hosted on your AWS account; the effort of running it was not measured