Understanding high availability and disaster recovery for Cloud Functions

IBM Cloud® Functions is deprecated. Existing Functions entities such as actions, triggers, or sequences will continue to run, but as of 28 December 2023, you can’t create new Functions entities. Existing Functions entities are supported until October 2024. Any Functions entities that still exist on that date will be deleted. For more information, see Deprecation overview.

Learn about the high availability features of the IBM Cloud® Functions and design your resources to meet the availability requirements that your business and customers need.

High availability is a core discipline in an IT infrastructure to keep your resources healthy and your app workloads up and running, even after a partial or a full site failure. The main purpose of high availability is to eliminate potential points of failure in an IT infrastructure.

Regions and service level: All IBM Cloud general availability (GA) services have a Service Level Agreement of 99.99% availability. Cloud Functions is a GA service that is offered in six regions: US South, US East, EU Central, Tokyo, UK South, and Sydney. Each location uses different availability zones for redundancy. If all the availability zones in a location fail, the Cloud Functions service for that location becomes unavailable.; Cloud Functions is available only on the public cloud.
Who is responsible to set up high availability for my resources?: Your global strategy is important, and you are responsible for understanding your configuration, customization, and usage of the service to best leverage the capabilities of IBM Cloud. For more information, see the topic How does IBM Cloud ensure zero downtime?; While Cloud Functions is responsible to ensure that your namespace information is available, backed up, and replicated across multiple regions so that information can be recovered after a failure, Cloud Functions does not set up high availability for any IBM Cloud resources that are used by Cloud Functions. Instead, you must understand the high availability options that each resource offering provides to implement the right level of availability for your needs.

See ensure zero downtime to learn more about the high availability and disaster recovery standards in IBM Cloud. You can also find information about Service Level Agreements.

The following sections contain information about high availability, disaster recovery for Cloud Functions as well as HIPAA-related considerations.

High availability

Cloud Functions is highly available within multiple IBM Cloud regions. Cloud Functions supports high availability to avoid a single point of failure. This service achieves high availability automatically and transparently through the multi-zone region (MZR) feature provided by IBM Cloud.

Disaster recovery

Customer data, such as action code, is automatically synchronized between database instances in different zones. In addition, IBM Cloud backs up your data by taking snapshots for use in data restoration in a disaster situation. If a disaster occurs, Cloud Functions switches over to use these snapshot backups automatically.

The likelihood of a region-wide disaster is low. However, if a disaster happens, IBM executes disaster-recovery plans to restore the service into another region. IBM Cloud® Functions then uses the regional backup of your data and the compute resources of another region to process your requests.

IBM Cloud® Functions regional failover locations
Location	Failover location
US South	US East
US East	US South
Tokyo	EU Central
UK South	EU Central
EU Central	Amsterdam
Sydney	Tokyo

Because Amsterdam does not have any active compute resources for IBM Cloud® Functions, the failover period in the EU Central region might be longer than in other regions.

Cloud Functions is a Function-as-a-Service offering, and as such, it offers only compute services. It does not back up the data that your workload (actions) processes. To store data, you can use separate cloud-based data stores with separate disaster recovery options, which depend on the storage service's SLAs.

HIPAA

Cloud Functions is a Function-as-a-Service offering, and as such, it offers only compute services. It does not offer storage options nor management of your data. Ensuring the HIPAA-compliant processing of your data is subject to your storage service's operation and SLAs.