Compute Engine SLAGoogle Cloud · Compute · Google Cloud · Compute
Promise vs reality
Downtime used against the allowance
3.7 h of major incidents, 7.3x the allowance30 min allowed in 208 days
1 major and 1 minor incidents; worst region global. 0 min degraded (not counted). History covers 208 of 365 days.
Major incident minutes per month
Allowed per month: 4.4 min
- Promised
- 99.99%
- Observed, 365d
- 99.927%
- Major incident time
- 3.7 h
- Credit on first breach
- 10%
Commitments and credits
- Instances in Multiple Zones (Premium Tier)VMs across two or more zones in a region; Cloud Regions excluding Mexico and Stockholm99.99%
Monthly uptime below, credit of the bill
- Below 99.99%10%
- Below 99%25%
- Below 95%100%
- Instances in Multiple Zones (Premium Tier, Mexico and Stockholm)Multi-zone VMs in the Mexico and Stockholm regions99.95%
Monthly uptime below, credit of the bill
- Below 99.95%10%
- Below 99%25%
- Below 95%100%
- Instances in Multiple Zones (Standard Tier network)Multi-zone VMs on the Standard Network Service Tier99.9%
Monthly uptime below, credit of the bill
- Below 99.9%10%
- Below 99%25%
- Below 95%100%
- Single Instance, Memory Optimized familyA single memory-optimized VM (regions excluding Mexico and Stockholm)99.95%
Monthly uptime below, credit of the bill
- Below 99.95%10%
- Below 95%25%
- Below 90%100%
- Single Instance, all other familiesA single VM of any other machine family (all regions)99.9%
Monthly uptime below, credit of the bill
- Below 99.9%10%
- Below 95%25%
- Below 90%100%
Terms
- Measured
- Calendar month, per Project per Region (per instance for a Single Instance), per Network Service Tier
- Credit cap
- Will not exceed the amount due for the respective Covered Services in the Regions that did not meet SLO for the month
- Claim window
- Within 60 days from the time Customer becomes eligible
- How to claim
- Notify Google technical support; provide log files/identifying info showing Downtime Periods and when they occurred
Not covered
- Features designated pre-general availability
- Features excluded from the SLA in the Documentation
- Errors caused by factors outside Google's reasonable control
- Errors from Customer's or third-party software or hardware
- Abuses or behaviors that violate the Agreement
- Errors resulting from quotas applied by the system / listed in the Admin Console
Premium Tier Cloud Regions (excluding Mexico and Stockholm) ... Instances in Multiple Zones >= 99.99% ... 99.00% - < 99.99% (Premium Tier) 10% ... 95.00% - < 99.00% 25% ... < 95.00% 100%
Load balancing has the same 99.99% (Premium) / 99.9% (Standard) SLO. Downtime = loss of external connectivity or persistent disk access.
Evidence: incidents, last 365 days
- Sep 1, 2026Multiple products in us-central1-b are experiencing network service degradation.us-central1Minor
- Aug 20, 2026We are investigating an issue where customers may experience timeouts, service degradations, errors, and elevated latencies across multiple products in the us-west1 region.global, us-west1Outage
SLA changes
- Aug29Compute EngineSingle Instance SLO raised from 99.5% to 99.9% (99.95% for memory-optimized family).Source
- Apr12Compute EngineSingle Instance SLO of 99.5% added (Jan 13, 2020 version had no single-instance commitment).Source
- Apr13Compute EngineCompute Engine SLO raised from 99.95% to 99.99%. Nov 30, 2017 version: 'Monthly Uptime Percentage to Customer of at least 99.95%'; April 13, 2018 version: 'at least 99.99%'.Source
Other compute SLAs
ServiceDowntime vs allowed
Amazon EC2AWS99.99% / 99.178%
Cloudflare WorkersCloudflare99.99% / 100%
Confluent Cloud for Apache FlinkConfluent99.99% / 99.628%
Droplets (CPU)DigitalOcean99.99% / 100%
Compute EngineGoogle Cloud99.99% / 99.927%
Azure FunctionsMicrosoft Azure99.95% / n/a
Promised in the SLAObserved on the vendor's status pagePartial historyShortfall
Summaries of published SLAs; the contract you sign governs. Logos via logo.dev; trademarks belong to their owners.