RunsOn m8a.xlarge · EBS gp3 vs Namespace 4x8 x64
RunsOn does the Rust build for 3.0× less; neither is clearly faster on the Rust build (within 5% or inside each other's run-to-run range). Both 4 vCPU x64.
Price per minute 4.6× cheaper on RunsOn: $0.0013 vs $0.0060the builds are not all compared on cost yet
Rust build 3.0× cheaper on RunsOn: $0.0034 vs $0.0102within run-to-run spread: 1m 49s · 1m 35s
TypeScript build not yet RunsOn 1m 03s (7 runs) · Namespace 45s (1 run)
Docker build 9m 08s RunsOn only · Namespace not run
RunsOn m8a.xlarge · EBS gp3
jobs ran from Amazon.com (US)
Namespace 4x8
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 | $0.0034 | $0.0102 | 3.0× RunsOn |
| Cost per TypeScript build | $0.0021 | $0.0052 | not yet |
| Cost per Docker build | $0.0162 | n/a: Not run: the PostHog Docker build needs more than 8 GB of memory; it ran out of memory on 8 GB runners. | — |
| Price per minute | $0.0013 | $0.0060 overage rate | 4.6× RunsOn |
| Real builds, wall time | |||
| Rust build (spotify-player) | 1m 49s | 1m 35s | ≈ even |
| TypeScript build (opencode) | 1m 03s | 45s | not yet |
| Docker build (PostHog) | 9m 08s | n/a: Not run: the PostHog Docker build needs more than 8 GB of memory; it ran out of memory on 8 GB runners. | — |
| CPU & memory | |||
| PassMark single-thread (points) | 3,809 | 4,659 | 1.2× Namespace |
| PassMark CPU Mark, all cores (points) | 13,481 | 16,112 | 1.2× Namespace |
| sysbench CPU, 1 thread (events/s) | 5,012 | 6,909 | 1.4× Namespace |
| 7-Zip compression (MIPS) | 38,485 | 33,838 | 1.1× RunsOn |
| Memory read bandwidth | 36.6 GiB/s | 45.0 GiB/s | 1.2× Namespace |
| Measurement | RunsOn | Namespace | Gap |
|---|---|---|---|
| Workspace disk (fio) | |||
| Random read, 4k QD32 (IOPS) | 2,994 | 1,209,187 | 403.9× Namespace |
| Random write, 4k QD32 (IOPS) | 2,991 | 607,540 | 203.1× Namespace |
| Random read, 4k QD1 (IOPS) | 1,825 | 105,744 | 58.0× Namespace |
| QD1 read latency, p99 | 807 µs | 10 µs | 78.0× Namespace |
| Sequential read | 401 MiB/s | 5.0 GiB/s | 12.8× Namespace |
| Sequential write | 401 MiB/s | 3.4 GiB/s | 8.8× Namespace |
| Sync write, 4k + fdatasync (IOPS) | 540 | 30,836 | 57.1× Namespace |
| Network, cache & Docker | |||
| Download from GitHub | 159 MiB/s | 128 MiB/s | 1.2× RunsOn |
| Cache restore, 4 GB (actions/cache) | 16s | 18s | 1.1× RunsOn |
| Cache save, 4 GB (actions/cache) | 16s | 14s | 1.2× Namespace |
| docker pull | 4.8s | 3.2s | 1.5× Namespace |
| Database | |||
| pgbench read/write (TPS) | 3,256 | 13,778 | 4.2× Namespace |
| pgbench read-only (TPS) | 142,376 | 114,046 | 1.2× RunsOn |
| Queue | |||
| Queue under burst load | 17s | 9s | 1.9× Namespace |
- ≈ even: under 5% apart, or a build inside both runners' run-to-run spread
- not yet: one side has fewer than 3 runs, so the two values are shown but not compared
- Queue: identical jobs queued at the same instant (burst suite)
- 2 shown, not compared yet (under 3 runs on a side)
- 2 missing on one side or both
- RunsOn at EC2 spot, its default (7-day average of the cheapest us-east-1 zone's spot price, fetched 16 Nov 2026); 9 of its jobs fell back to on-demand (AWS had no spot capacity for the type) and are priced at the spot rate over the same billed time
- Cost: a managed SaaS runner against RunsOn self-hosted on your AWS account; the effort of running it was not measured