All services / Reliability / November 2026
Cloud & developer reliability leaderboard: November 2026
Live: month to date (2 days so far), updated hourly. Results freeze on the 1st. Ranked by uptime from our own checks, then fewest outages, then least downtime. Methodology · JSON
Most reliable
- Akamai · 100%
- AWS · 100%
- Bitbucket · 100%
- Datadog · 100%
- DigitalOcean · 100%
- Docker Hub · 100%
- Fastly · 100%
- Firebase · 100%
- GitLab · 100%
- Google Cloud · 100%
Most outage-prone
- Sentry · 1 outage, 1 h 55 min down · 95.37%
- Fly.io · 1 outage, 36 min down · 98.55%
- Cloudflare · 1 outage, 30 min down · 98.80%
Full ranking (32)
| # | Service | Uptime | Outages | Total downtime | Median outage | User reports |
|---|---|---|---|---|---|---|
| 1 | Akamai | 100% | 0 | — | — | 0 |
| 2 | AWS | 100% | 0 | — | — | 0 |
| 3 | Bitbucket | 100% | 0 | — | — | 0 |
| 4 | Datadog | 100% | 0 | — | — | 0 |
| 5 | DigitalOcean | 100% | 0 | — | — | 0 |
| 6 | Docker Hub | 100% | 0 | — | — | 0 |
| 7 | Fastly | 100% | 0 | — | — | 0 |
| 8 | Firebase | 100% | 0 | — | — | 0 |
| 9 | GitLab | 100% | 0 | — | — | 0 |
| 10 | Google Cloud | 100% | 0 | — | — | 0 |
| 11 | Heroku | 100% | 0 | — | — | 0 |
| 12 | Microsoft Azure | 100% | 0 | — | — | 0 |
| 13 | MongoDB Atlas | 100% | 0 | — | — | 0 |
| 14 | Netlify | 100% | 0 | — | — | 0 |
| 15 | npm | 100% | 0 | — | — | 0 |
| 16 | Okta | 100% | 0 | — | — | 0 |
| 17 | Oracle Cloud | 100% | 0 | — | — | 0 |
| 18 | PyPI | 100% | 0 | — | — | 0 |
| 19 | Render | 100% | 0 | — | — | 0 |
| 20 | Salesforce | 100% | 0 | — | — | 0 |
| 21 | Squarespace | 100% | 0 | — | — | 0 |
| 22 | Stack Overflow | 100% | 0 | — | — | 0 |
| 23 | Supabase | 100% | 0 | — | — | 0 |
| 24 | Twilio | 100% | 0 | — | — | 0 |
| 25 | Vercel | 100% | 0 | — | — | 0 |
| 26 | Wix | 100% | 0 | — | — | 0 |
| 27 | WordPress.com | 100% | 0 | — | — | 0 |
| 28 | GitHub | 99.73% | 0 | — | — | 0 |
| 29 | Jira | 99.66% | 0 | — | — | 0 |
| 30 | Cloudflare | 98.80% | 1 | 30 min | 30 min | 0 |
| 31 | Fly.io | 98.55% | 1 | 36 min | 36 min | 0 |
| 32 | Sentry | 95.37% | 1 | 1 h 55 min | 1 h 55 min | 0 |
Methodology
- Uptime: share of the covered time a service was up. Downtime is the time covered by our own failed HTTP checks (each failed check counts for one check interval; checks run from Cloudflare's network) merged with confirmed outages (below), even ones our checks missed. Overlapping periods are counted once. It measures reachability, not every feature.
- Outages: public outages in our outage log (outage level or worse for at least 10 minutes, or a major outage), excluding ones flagged as false positives. Outages are confirmed by the vendor's status page and/or user reports. Downtime is clipped to the period.
- User reports: problem reports submitted on this site during the period.
- Minimum data: a service is ranked only with checks on at least 60% of the days in the period (2 of 2) and at least 24 checks (about 12 a day). Services ranked with fewer than 50 checks a day are marked lower confidence (*); short outages between their checks are more likely to be missed. 730 services were left out for too little data.
- Order: uptime (high to low), then fewer outages, less downtime, fewer reports. Month pages freeze on the 1st (UTC); the current month is live.