Service Level Agreement (SLA) - EDB Postgres AI Cloud Services

Introduction

This Service Level Agreement (“SLA”) sets forth the applicable service levels and service level credits for Customer’s use of EDB Postgres AI Cloud Services. The SLA service level credits are the Customer’s exclusive remedy for EDB's failure to meet a specified service level.

This SLA is incorporated into the EDB EULA (the “Agreement”).

General Terms

Definitions 

All terms capitalized will have the definition provided in the Agreement unless otherwise specified herein. 

“Beta” or “Tech Preview” refers to any service or feature that has been made available in pre-production release or on a testing basis.

“Cluster(s)” refers to the following types of clusters: 

• “Single Node Cluster” is a cluster without high availability enabled. 
• “HA Cluster” or “High Availability Cluster” refers to a cluster with high availability enabled, not powered by EDB Postgres AI Distributed. 
• “PGD Cluster” refers to a cluster powered by EDB Postgres AI Distributed, which may contain one or more Data Groups in a single or multiple regions. 

 “Data Group” refers to a group of nodes in a single Region that are part of a PGD Cluster.

“Database Uptime” or “Uptime” is calculated in the following ways: 

• A minute is considered available if either: (a) there are no connections to the Cluster or Faraway Replica, or no active connections issuing queries; (b) there is at least one successful client connection established; or (c) an active connection issues a query to the Cluster or Faraway Replica. 
• Minutes during which the Cluster or Faraway Replica is undergoing scheduled maintenance or upgrades are considered available. 
• Single Node & HA Clusters or Faraway Replica availability is the total number of minutes during the month that the Cluster or Faraway Replica is available. 
• PGD Cluster availability is the total number of minutes during the month that all Data Groups in the PGD Cluster are available.

“Downtime” is calculated per Cluster or Faraway Replica on a monthly basis, and is the total number of minutes during the month that the Cluster or Faraway Replica is unavailable. 

“Faraway Replica” refers to the faraway replica feature in EDB Postgres AI. Faraway replicas are read-only replicas of EDB Postgres AI Single Node or High Availability Clusters that you can provision in most supported Regions. Database users and applications can read from replicas instead of the source Cluster. 

“Region” refers to the geographic area as defined by the applicable cloud service provider from which you deploy EDB Postgres AI, as the case may be.

“Service Credit” is calculated as a percentage of the charges paid by you for EDB Postgres AI for the month in the billing cycle in which the applicable SLA was not met.

“Service Level” refers to the applicable monthly Uptime percentage availability of a specified Cluster. 

“Support” means the technical support and maintenance services as described in the then-current EDB Support Policy.

“Cloud Service Provider” or “CSP” is the infrastructure provider your databases live on - for example Amazon’s AWS or Microsoft's Azure platforms. 

Claims

In order for EDB to consider a claim for Service Credit, you must submit the claim within the prescribed time frame to the EDB Support Team at: cloudsupport@enterprisedb.com including all necessary information for EDB to validate the claim.

To be eligible for a Service Credit you must: 

1. Submit a support ticket with the Support Team within twenty-four (24) hours of becoming aware of an event that impacts service availability. 
2. Submit your claim for a Service Credit within sixty (60) days of the event that impacted service availability. 
3. Include all information necessary to validate your Service Credit request within your claim, including (i) a detailed description of the events resulting in Downtime, including your logs that document errors and corroborate the claimed outage, with confidential information redacted; (ii) the time and duration of Downtime; (iii) the number and locations of affected users as applicable; (iv) details regarding  your attempts to resolve the Downtime at the time it occurred. 
4. Reasonably assist the Support Team in investigating the cause of Downtime to process your claim.
5. Comply EDB documentation and guidance from the EDB Support Team.
    

 In the event that the EDB Postgres Cloud Service does not meet the 99.995% Uptime availability, you may receive a Service Credit as set forth below:

SLA for PGD Clusters with Data Groups in Multiple Regions

 

SLA for HA Clusters and PGD Clusters with Data Group(s) in a Single Region

 
SLA for Single Node Clusters and Faraway Replicas

 Service Credits are your sole remedy for any performance or availability issues and only apply to fees paid for the Cluster(s) or Faraway Replica(s) impacted for which a Service Level was not met. Service Credits are for future use of service and are capped at 100% of paid fees in the month in question. 

Limitations

1. You will not be eligible for a Service Credit for any performance or availability issue that results from:
   Factors outside of our reasonable control, such as natural disaster, war, acts of terrorism, riots, government action, or a network or device failure between your client application and EDB Postgres Cloud Service;
2. Services, hardware, or software provided by a third party, such as cloud platform services on which EDB Postgres AI runs;
3. Use of your password or equipment to access the EDB Postgres AI network;
4. Your or any third party’s (a) improper use, scaling or configuration of EDB Postgres AI, or (b) failure to follow appropriate security practices; or
5. Your or any third party’s tampering with cloud platform services, hardware, or software managed by EDB Postgres AI; or
6. Periods during which EDB Postgres AI has scheduled maintenance; or
7. EDB Postgres AI Beta or Tech Preview Offerings.