Fastest GitHub Actions runners: disk I/O.

How this shows up in real builds →

What the numbers say

Extremes on x64, and RAM against disk.

• On x64, one small read at a time (4k, QD1) comes back 85,326 times a second on Blacksmith 4 vCPU and 1,569 on RunsOn m8i.2xlarge · EBS gp3, 54× apart; p99 latency is 14 µs against 1,012 µs.
• Synced 4k writes on x64 run from 495 per second on RunsOn m8i.2xlarge · EBS gp3 (workspace mounted nobarrier) to 20,425 on Blacksmith 4 vCPU (workspace mounted nobarrier).
• r8a.xlarge with tmpfs ($0.0021/min at spot) answers 1,071,527 QD1 reads and 431,794 synced writes per second: 88× and 36× the best local NVMe at 4 vCPU (i7i.xlarge, $0.0019) and 586× and 800× default EBS gp3 (m8a.xlarge, $0.0015); its workspace is 31 GiB, from RAM. (EC2 storage suite)

More findings20: EBS, arm64, NVMe vs EBS, provisioned gp3, tmpfs, storage classes

• Every EBS-backed workspace (3 AWS CodeBuild x64 runners, 25 RunsOn configurations and 3 Warpbuild configurations) answers between 1,569 and 1,844 QD1 reads and 495 to 562 synced writes per second, whoever runs it and whatever throughput it is provisioned for: each request is a round trip to network storage.
• On arm64, one small read at a time (4k, QD1) comes back 26,866 times a second on Warpbuild xfast (Apple M4 Pro) 6 vCPU and 1,660 on RunsOn m9g.large · EBS gp3, 16× apart; p99 latency is 40 µs against 963 µs.
• Synced 4k writes on arm64 run from 537 per second on RunsOn c8g.large · EBS gp3 (workspace mounted nobarrier) to 17,054 on RunsOn m9gd.2xlarge · local NVMe.
• Same Xeon 6975P-C CPU, different disk: m8id.2xlarge on local NVMe answers 11,063 QD1 reads and 7,665 synced writes per second, m8i.2xlarge on EBS gp3 1,569 and 495.
• Same Xeon 6975P-C CPU, different disk: m8id.large on local NVMe answers 10,991 QD1 reads and 5,549 synced writes per second, m8i-flex.large on EBS gp3 1,745 and 528.
• Same Xeon 6975P-C CPU, different disk: m8id.xlarge on local NVMe answers 11,033 QD1 reads and 7,853 synced writes per second, m8i-flex.xlarge on EBS gp3 1,765 and 527.
• Same Neoverse-V3 CPU, different disk: m9gd.2xlarge on local NVMe answers 13,522 QD1 reads and 17,054 synced writes per second, m9g.2xlarge on EBS gp3 1,788 and 543.
• Same Neoverse-V3 CPU, different disk: m9gd.xlarge on local NVMe answers 11,717 QD1 reads and 11,264 synced writes per second, m9g.xlarge on EBS gp3 1,713 and 543.
• Same Neoverse-V3 CPU, different disk: m9gd.large on local NVMe answers 13,477 QD1 reads and 7,176 synced writes per second, m9g.large on EBS gp3 1,660 and 542.
• Provisioning gp3 at 1000 MiB/s on m8a.2xlarge takes sequential reads from 402 MiB/s to 1,002 MiB/s (2.5×); QD1 reads barely move (1,791 → 1,792 IOPS): it buys bandwidth, not latency.
• Provisioning gp3 at 1000 MiB/s on m8azn.xlarge takes sequential reads from 401 MiB/s to 1,001 MiB/s (2.5×); QD1 reads barely move (1,775 → 1,775 IOPS): it buys bandwidth, not latency.
• tmpfs on m8a.2xlarge is not a disk at all: 1,156,332 QD1 reads and 432,842 synced writes per second, from RAM. The workspace is 31 GiB, sized by memory.
• tmpfs on m9g.2xlarge is not a disk at all: 653,143 QD1 reads and 301,463 synced writes per second, from RAM. The workspace is 31 GiB, sized by memory.
• tmpfs on m9g.xlarge is not a disk at all: 680,179 QD1 reads and 303,962 synced writes per second, from RAM. The workspace is 15 GiB, sized by memory.
• tmpfs on r8a.2xlarge is not a disk at all: 1,151,153 QD1 reads and 433,820 synced writes per second, from RAM. The workspace is 62 GiB, sized by memory.
• tmpfs: 653k–1.2M QD1 reads and 301k–434k synced writes per second across 5 runners (5 RunsOn configurations).
• VM disk: 5.9k–85k QD1 reads and 578–20k synced writes per second across 42 runners from 7 providers (6 Avrea runners, 6 Blacksmith runners, 3 GitHub runners, 11 Namespace runners, 2 StarSling x64 runners, 9 Ubicloud runners and 5 Warpbuild runners).
• local NVMe: 11k–14k QD1 reads and 5.5k–17k synced writes per second across 9 runners (9 RunsOn configurations).
• network disk: 6.8k–8.6k QD1 reads and 3.3k–4.7k synced writes per second across 2 runners (2 GitHub runners).
• EBS: 1.6k–1.8k QD1 reads and 495–562 synced writes per second across 31 runners from 3 providers (3 AWS CodeBuild x64 runners, 25 RunsOn configurations and 3 Warpbuild configurations).

Reading these numbers

Most of a build is small, dependent I/O, not one big stream.

Why latency before throughput

• git checkout, npm install, cargo build and test runners read thousands of small files, each waiting for the last: the QD1 column. QD32 and sequential show what a parallel compiler, a big docker build or a cache archive can pull; read them after.
• Package managers, databases and SQLite-backed tests flush their writes. Each flush waits for the storage: network disks pay a round trip every time. That is sync write.

Every runner

Each provider's best disk first; its other machines sit behind +N more. Bars are log scale.

Pick the disk per job on RunsOn

• The instance type sets the storage under the workspace: EBS gp3 by default, provisioned gp3 for bandwidth, local NVMe on the d and i families, or tmpfs when the job fits in RAM. Pick it with the family label; each one, measured: EC2 storage benchmark.

How it's measured

Methodology →

fio runs inside the job, in GITHUB_WORKSPACE: the disk a build writes to, on the filesystem the provider mounted there. Every runner is compared, whatever its shape.

fio settingsblock sizes, queue depths, mounts

• QD1 random read: 4k blocks, one request in flight; IOPS and p99 latency.
• Sync write: 4k writes, each followed by fdatasync.
• Random read and write: 4k blocks, 4 jobs with 32 requests in flight each.
• Sequential read and write: 1M blocks, 32 requests in flight.
• The workspace filesystem, its size and its mount options come from the job's own mount table.
• Earlier versions of this page used a different harness and other fio settings on 2 vCPU runners; those numbers are gone rather than mixed in.