Uptimeby GlyphSnowflake Service (Data Cloud) SLA
Snowflake · Data platforms · promises up to 99.99% · SLA of Feb 24, 2026 · checked 47d ago
30 major outages in the last 365 days. Worst region Azure - East US 2 (Virginia): 5.1 d counted, 98.605% observed against the 99.99% Service Level at 10% Error Rate Threshold, which applies per region.
9 of 12 months fell below it (credits are claimed per month). One incident ran past 72 hours; each counts for 72 hours at most (see the method).
Last 90 days: 8.9 h counted, 99.589% observed.
Incidents, last 365 days
Azure - West US 2 (Washington), Azure - South Central US (Texas)
A minor incident for 3.4 h, not counted as downtime (ticket INC20000321).
10 days ago
Azure - West US 2 (Washington), Azure - South Central US (Texas)A minor incident for 21 h, not counted as downtime (ticket INC20000319).
11 days ago
AWS - Europe (Ireland)A major outage for 5 h, counted as downtime (ticket INC20000301).
2 weeks ago
Azure - West US 2 (Washington)A minor incident for 51 min, not counted as downtime (ticket INC20000299).
2 weeks ago
GCP - Europe West 2 (London)A major outage for 8.9 h, counted as downtime (ticket INC20000294).
2 weeks ago
AWS - Asia Pacific (Seoul)A minor incident for 44 min, not counted as downtime (ticket INC20000290).
3 weeks ago
GCP - Middle East (Dammam)A major outage for 29 min, counted as downtime (ticket INC20000287).
3 weeks ago
AWS - US East (Ohio)A minor incident for 2.2 d, not counted as downtime (ticket INC20000277).
3 weeks ago
Azure - UAE North (Dubai), GCP - Middle East (Dammam)A partial outage or slowdown for 73 min, not counted as downtime (ticket INC20000263).
4 weeks ago
AWS - US West (Oregon), AWS - US West (Commercial Gov - Oregon)A major outage for 61 min, counted as downtime (ticket INC20000239).
6 weeks ago
AWS - Europe (Ireland)A major outage for 80 min, counted as downtime (ticket INC20000237).
6 weeks ago
AWS - Asia Pacific (Sydney), AWS - Asia Pacific (Singapore)A partial outage or slowdown for 81 min, not counted as downtime (ticket INC20000217).
8 weeks ago
AWS - Europe (Frankfurt), AWS - Asia Pacific (Tokyo)A major outage for 6 h, counted as downtime (ticket INC20000213).
8 weeks ago
AWS - US East (N. Virginia)A major outage for 2 h, counted as downtime (ticket INC20000211).
2 months ago
AWS - Europe (Frankfurt)A major outage for 2.6 h, counted as downtime (ticket INC20000199).
2 months ago
AWS - US West (Oregon), AWS - GovCloud (US-West)A partial outage or slowdown for 3.9 h, not counted as downtime (ticket INC20000190).
2 months ago
AWS - US East (N. Virginia)A major outage for 2.4 h, counted as downtime (ticket INC20000188).
2 months ago
GCP - US Central 1 (Iowa)A minor incident, still open, not counted as downtime (ticket INC20000150).
72 h counted, ran 74 d2 months ago
Azure - Central US (Iowa), Azure - Central India (Pune)A partial outage or slowdown for 81 min, not counted as downtime (ticket INC20000182).
2 months ago
Azure - Southeast Asia (Singapore)A major outage for 42 min, counted as downtime (ticket INC20000175).
2 months ago
AWS - US West (Oregon)A partial outage or slowdown for 1.7 h, not counted as downtime (ticket INC20000173).
2 months ago
Azure - East US 2 (Virginia)A partial outage or slowdown for 1.9 h, not counted as downtime (ticket INC20000168).
2 months ago
AWS - US East (N. Virginia)A partial outage or slowdown for 1.9 h, not counted as downtime (ticket INC20000163).
2 months ago
Azure - East US 2 (Virginia)A major outage for 2.1 h, counted as downtime (ticket INC20000158).
2 months ago
AWS - US East (Ohio)A major outage for 69 min, counted as downtime (ticket INC20000157).
2 months ago
Latest 25 of 60.
Major outage time per month
Allowed per month: 4.4 min
Commitments and credits
Service Level at 10% Error Rate Threshold
All customers of the Service (excluding Features); same SLA regardless of edition
Credit: No credit schedule published
Service Level at 1% Error Rate Threshold
All customers of the Service (excluding Features); same SLA regardless of edition
Credit: No credit schedule published
SLA changes
Snowflake Service (Data Cloud)
The SLA page text changed (14 sentences added, 3 removed); it still states 99.99%, 99.95% and 99.9%.
2 weeks agoPage
Snowflake Service (Data Cloud)
Base support tier renamed from 'Premier Support' to 'Snowflake Support'; response times unchanged (Sev1 1 hour / Priority 15 minutes).
1 year agoSource
Snowflake Service (Data Cloud)
Single 99.9% Service Level (1% error threshold) replaced by dual definition adding 99.99% at a 10% Error Rate Threshold. Dec 10, 2021 version: 'The Monthly Availability Percentage for the Snowflake Service is ninety-nine and nine-tenths percent (99.9%)'; June 1, 2022 version: 'Monthly Availability Percentage of 99.99% when measured with a 10% Error Rate Threshold'. Credit multiples (1x/3x/7x) unchanged.
4 years agoSource
More in data platforms
Databricks publishes no SLA for it; its status page has no history this site can read.
Google Cloud promises 99.99% uptime, with a 10% credit when it misses; its status page showed 3.7 h of major outage in the 252 days on file.
Observed 99.939% over 365 days
Palantir publishes no SLA for it; its status page has no history this site can read.
Databricks Data + AI PlatformDatabricks publishes no SLA for it; its status page has no history this site can read.NoneFirst creditn/aObserved 365dNo feed
BigQueryGoogle Cloud promises 99.99% uptime, with a 10% credit when it misses; its status page showed 3.7 h of major outage in the 252 days on file.99.99%First credit10%Observed 365d99.939%All incidents 99.467%
Palantir Foundry and AIPUnverifiedPalantir publishes no SLA for it; its status page has no history this site can read.NoneFirst creditn/aObserved 365dNo feed
Terms, exclusions and sources
- Measured
- Monthly Availability Percentage per Cloud Provider Region; a minute is Unavailable when the Error Rate across all Accounts in the region exceeds the Error Rate Threshold (10% or 1%). Two Service Level definitions apply; customer picks the combination giving the larger credit.
- Claim window
- Within twenty-one (21) days of the calendar month in which the Service Level Failure occurred
- How to claim
- Customer requests Service Level Credits (support case); credited against usage in the affected Cloud Provider Region the following month
- Not covered
- Customer not processing data per Documentation and Snowflake recommendations; Equipment/software not provided by Snowflake (External Offerings) and third-party acts (cloud providers are NOT excluded); Force majeure; Evaluation, free trial, proof-of-concept or Preview-Terms features; Customer violation of the Agreement or suspension/termination; Snowpark Container Services scheduled maintenance and critical upgrades
- Status history
- Status history since Sep 22, 2025
- Weighted uptime
- All incidents, weighted: 99.991% · Downtime
Under 99.99% but greater than or equal to 99.0%* | 10% | 1 x Average Daily Snowflake Credits
Credits are not a % of the bill: they are multiples of 'Average Daily Snowflake Credits' (1x, 3x, 7x), so the schema's numeric credit field is null. Schedule: <99.99%/10% threshold down to 99.0% = 1x; <99.9%/1% threshold down to 99.0% = 1x; <99.0% to 95.0% = 3x; <95.0% = 7x. Credits cannot be converted to money. The page makes no distinction by edition (Standard/Enterprise/Business Critical); no edition-specific SLA was found. Uptime counts both Unavailable Minutes (error-rate based) and Downtime (UI inaccessible due to server-side errors).