RunsOn m8azn.xlarge · gp3 1000 MiB/s vs GitHub 4-core x64
RunsOn does each build for 3.4–4.0× less and 1.9–2.6× faster. Both 4 vCPU x64.
All three builds 3.9× cheaper on RunsOn: $0.0645 vs $0.2498RunsOn 2.0× faster: 9m 21s vs 18m 27s
Rust build 3.4× cheaper on RunsOn: $0.0136 vs $0.0466RunsOn 2.0× faster: 1m 54s vs 3m 47s
TypeScript build 3.7× cheaper on RunsOn: $0.0072 vs $0.0268RunsOn 2.6× faster: 45s vs 1m 56s
Docker build 4.0× cheaper on RunsOn: $0.0437 vs $0.1764RunsOn 1.9× faster: 6m 42s vs 12m 44s
RunsOn m8azn.xlarge · gp3 1000 MiB/s
jobs ran from Amazon.com (US)
GitHub 4-core
jobs ran from Microsoft (US, CA)
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 | GitHub | Gap |
|---|---|---|---|
| Cost | |||
| Cost per Rust build | $0.0136 | $0.0466 | 3.4× RunsOn |
| Cost per TypeScript build | $0.0072 | $0.0268 | 3.7× RunsOn |
| Cost per Docker build | $0.0437 | $0.1764 | 4.0× RunsOn |
| Price per minute | $0.0047 | $0.0120 | 2.5× RunsOn |
| Real builds, wall time | |||
| Rust build (spotify-player) | 1m 54s | 3m 47s | 2.0× RunsOn |
| TypeScript build (opencode) | 45s | 1m 56s | 2.6× RunsOn |
| Docker build (PostHog) | 6m 42s | 12m 44s | 1.9× RunsOn |
| CPU & memory | |||
| PassMark single-thread (points) | 4,257 | 2,370 | 1.8× RunsOn |
| PassMark CPU Mark, all cores (points) | 15,102 | 6,130 | 2.5× RunsOn |
| sysbench CPU, 1 thread (events/s) | 6,220 | 3,670 | 1.7× RunsOn |
| 7-Zip compression (MIPS) | 31,030 | 16,137 | 1.9× RunsOn |
| Memory read bandwidth | 40.8 GiB/s | 20.9 GiB/s | 2.0× RunsOn |
| Measurement | RunsOn | GitHub | Gap |
|---|---|---|---|
| Workspace disk (fio) | |||
| Random read, 4k QD32 (IOPS) | 15,994 | 19,587 | 1.2× GitHub |
| Random write, 4k QD32 (IOPS) | 15,991 | 11,712 | 1.4× RunsOn |
| Random read, 4k QD1 (IOPS) | 1,775 | 6,843 | 3.9× GitHub |
| QD1 read latency, p99 | 856 µs | 251 µs | 3.4× GitHub |
| Sequential read | 1,002 MiB/s | 394 MiB/s | 2.5× RunsOn |
| Sequential write | 1,001 MiB/s | 394 MiB/s | 2.5× RunsOn |
| Sync write, 4k + fdatasync (IOPS) | 540 | 3,262 | 6.0× GitHub |
| Network, cache & Docker | |||
| Download from GitHub | 256 MiB/s | 91 MiB/s | 2.8× RunsOn |
| Cache restore, 4 GB (actions/cache) | 10s | 41s | 4.0× RunsOn |
| Cache save, 4 GB (actions/cache) | 14s | 30s | 2.1× RunsOn |
| docker pull | 5.0s | 16.4s | 3.3× RunsOn |
| Database | |||
| pgbench read/write (TPS) | 3,758 | 2,797 | 1.3× RunsOn |
| pgbench read-only (TPS) | 163,575 | 20,013 | 8.2× RunsOn |
| Queue | |||
| Queue under burst load | 20s | 4s | 5.0× GitHub |
- ≈ 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 5 Oct 2026); the 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