Google has recently announced the preview of Backup for GKE, a cloud-native way to protect, manage, and restore containerized applications and data running on Kubernetes.
Using the new service developers can create backups plans to schedule periodic backups of both application data and GKE cluster state data, restore each backup to a cluster in the same or in a different region.
With the increased adoption of Kubernetes and containerized applications, more workloads, including relational databases, require managed backup solutions. Guru Pangal, GM storage at Google Cloud, and Brian Schwarz, director of product management, write:
Google Cloud users continue to adopt GKE in droves (..) and they’re no longer just running stateless applications in containers; they’re also running databases such as MySQL and PostgreSQL inside containers, as well as other stateful workloads.
Prior to Backup for GKE, many GKE customers backed up their stateful application data separately from GKE cluster state data. Application data could be protected via a storage-based backup, while cluster state data might be captured occasionally using custom scripts and stored in a separate customer bucket. Customers with ongoing backup requirements relied on homegrown solutions to perform regular backups and to demonstrate compliance (…) Storage management tasks, like creating a clone for testing purposes, or migrating data from one cluster to another, meant additional operational overhead.
Even if the cloud provider documented how persistent disk features could be used with GKE, the lack of a native way to backup GKE clusters was a challenge for many developers. In a Reddit thread, users are suggesting different approaches that require significant coding and testing, third party tools, or do not support multi-region deployments.
Daniel Dersch, senior manager cloud architecture at Mitel, questions instead if Backup for GKE can be used to extend the service range of a cluster:
So say we accidentally provisioned a GKE cluster a number of months ago, but we didn’t provide enough IP addresses for the “Service Range”. Could we feasibly backup the cluster, nuke it, and recreate it with a bigger service range? Asking for a friend…
Besides Backup for GKE, Pangal and Schwarz announced other new storage features for Google Cloud: Filestore Enterprise, introducing high availability via synchronous replication across multiple zones, and custom region selection and optional 15 min RPO for dual-region buckets.
The pricing of Backup for GKE has not been announced yet. To access the preview, existing customers should contact their account team or sales representatives.