Kubernetes Cluster Backup Using Persistent Volume Snapshots

The patent segments the service cluster instance into multiple container groups, and further segments the backup process into distinct phases (snapshot creation, data copying, verification). By introducing persistent volume claims and persistent volume objects as intermediate management layers, the system divides the complex backup task into manageable units that can be processed independently and in parallel, thereby improving backup efficiency without overwhelming the system.

The patent performs preliminary actions by pre-configuring persistent volume claims and persistent volume objects before the actual backup process. Storage resources are pre-allocated and bound to container groups in advance, and snapshot mechanisms are prepared beforehand. This preliminary setup enables the backup process to proceed rapidly without ad-hoc resource allocation during the critical backup window, thus improving productivity while maintaining controlled complexity.

The patent employs copying mechanisms through persistent volume objects that create references to storage resources without immediately duplicating all data. Snapshot functionality creates point-in-time copies of the storage state, and data copying is performed selectively based on changes detected since the last backup. This approach ensures data completeness while dramatically reducing the time required compared to traditional full backup methods.

The patent replaces mechanical data copying with more efficient storage-level mechanisms. Instead of application-level data extraction and transfer, the system uses persistent volume snapshots and storage area network (SAN) copying capabilities. This substitution leverages hardware and storage system optimizations to achieve rapid, complete backups without the overhead of traditional software-based backup processes.