---
title: "Dev Infrastructure Uptime & Reliability"
description: "30-day and 90-day uptime rankings for the largest Dev Infrastructure providers. Independent measurement, no provider affiliation."
url: https://uptimeproject.org/category/dev-infra/
built: 2026-11-06T12:58:15.145Z
---

# Dev Infrastructure Reliability Rankings

All 10 Dev Infrastructure providers ranked by 30-day availability. Independent measurement, no provider affiliation. Refreshed on every page load.

Dev Infrastructure services ranked by 30-day availability
| Rank | Provider | Status | 7d | 30d availability | 30d downtime | 90d |
| --- | --- | --- | --- | --- | --- | --- |
| Tied for 1 | [crates.io](https://uptimeproject.org/service/crates-io/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| Tied for 1 | [Docker Hub](https://uptimeproject.org/service/docker-hub/) | Operational | 100.00% | 100.00% | 0 min | 99.99% |
| Tied for 1 | [GitHub](https://uptimeproject.org/service/github/) | Operational | 100.00% | 100.00% | 0 min | 99.92% |
| Tied for 1 | [GitHub Container Registry](https://uptimeproject.org/service/ghcr/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| Tied for 1 | [Go Module Proxy](https://uptimeproject.org/service/go-module-proxy/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| Tied for 1 | [Maven Central](https://uptimeproject.org/service/maven-central/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| Tied for 1 | [npm Registry](https://uptimeproject.org/service/npm-registry/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| Tied for 1 | [PyPI](https://uptimeproject.org/service/pypi/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| Tied for 1 | [RubyGems.org](https://uptimeproject.org/service/rubygems/) | Operational | 100.00% | 100.00% | 0 min | 100.00% |
| #10 | [GitLab](https://uptimeproject.org/service/gitlab/) | Operational | 100.00% | 99.95% | 12 min+6 min degraded | 99.94% |

## What does this measure?

Independent probes in five locations measure each service at per-check cadences: HTTP every minute, DNS every five minutes, TLS once an hour, storage every five to ten minutes. A minute is "up" only when a strict majority of reporting probes vote up; minutes with fewer than three reporting probes are excluded from the denominator. The 30-day availability above is the share of measured minutes flagged "up". Read the [full methodology](https://uptimeproject.org/methodology/) .