Fio

Fio is a tool used to measure a computer's storage performance.

  • Development
  • Tools & Utilities
Fio is a tool used to measure a computer's storage performance. It simulates different workloads to test how quickly and reliably your hard drives or SSDs can read and write data. It helps ensure your hardware can handle the demands of your specific applications and tasks.

Platform availability

Available on 1 of 10 platforms

  • Unraid (not listed)
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows
  • Linux (not listed)
  • Any server (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

74/100 Good

Maintained
100last commit 4 days ago
Popular
766,433 GitHub stars
Easy to install
251 install platform, install notes
Light to run
No data
Security check
71no serious findings, 1 minor, CI, tests, OpenSSF Scorecard 4.5/10

Based on 4 of 5 factors.

GitHub stars
6,433
Open issues
246
Last commit
2026-11-17
Latest release
fio-3.43
License
GPL-2.0
Activity
rising

Checked 3 days ago - source: GitHub (axboe/fio)

Resources & Compatibility

Requirements not published yet.

First-install notes

  • Point fio at a safe test file Use --filename=mydata.bin and --size=100M so fio works on a scratch file and uses only the space you name. If you skip these, the job may fail, hang, or fill more of your disk than you expect. source
  • Only open the server port if needed If you run fio --server, it opens a network port (a numbered doorway for remote commands) on default port 8765. Keep that port closed or limited to your test client, or anyone on the network could start I/O tests on your machine. source
  • Try a short job before a long test Start with a small --size and a short --runtime so you can confirm the file path, permissions, and command work before a long run. Without limits, a bad job can hog your disk, fill a volume, or make you wait a long time. source

Fetched 53 days ago - each note links to its source

Alternatives (4)

  • Blackmagic Disk Speed Test - Disk Speed Test is an easy to use tool to quickly measure and certify your disk performance for working with high quality video!
  • CrystalDiskMark - Efficient tool for measuring HDD and SSD performance through sequential and random read/write speed tests, featuring a user-friendly interf…
  • KDiskMark - A simple open-source disk benchmark tool for Linux distros.
  • UserBenchMark - UserBenchmark is a free all-in-one benchmarking application.

Checked 54 days ago - source: AlternativeTo

Common questions

How do I keep fio from writing to an unintended file or filling the disk?

Point fio at a safe test file with --filename=mydata.bin and --size=100M so it works only on that scratch file. If you skip these, the job may fail, hang, or fill more of your disk than expected.

How should I restrict access when running fio in server mode?

Running fio --server opens a network port on default port 8765 for remote commands. Keep that port closed or limited to your test client, or anyone on the network could start I/O tests on your machine.

If fio reports I/O errors on Windows, can I fix it with a fio configuration change?

No, there is no fio configuration change that will help because the behavior depends on how your filesystem is configured. With regular ioengines, fio is at the mercy of the filesystem or block device for how long a sent I/O can be waited for.

How are global fio parameters applied to multiple jobs?

fio takes global parameters, and each thread inherits those settings unless a job provides parameters that override them.

Answers sourced from github.com, github.com, github.com, github.com

Community

Collected 48 days ago - public community threads