Bacula Systems, a player in high-security, high-performance backup and recovery solutions, announced the expansion of its HPC data protection portfolio with the introduction of its new BsnapDiff technology for ZFS and GPFS environments, alongside its Lustre Changelog integration technology.The new capabilities build upon Bacula’s recent investments in HPC-focused technologies, including its HPCAccelerator feature and native integrations with leading high-performance file systems. Together, these technologies are designed to help HPC architects, research institutions, large-scale enterprises, AI environments, and supercomputing facilities overcome one of the most persistent challenges in data protection: efficiently backing up environments containing hundreds of millions or even billions of files.

Traditional backup applications typically perform a complete scan of the file system during incremental backups to identify changed files. As data volumes and file counts continue to grow, this approach can create significant performance bottlenecks, increased metadata activity, extended backup windows, and unnecessary workload on production storage systems.

Bacula’s new BsnapDiff technologies take a different approach. By leveraging native file system capabilities, Bacula can identify changed files directly from file system metadata, snapshots, APIs, or changelogs, dramatically reducing or eliminating the need for full directory tree scans.

“Large-scale infrastructures require data protection technologies that work with the file system rather than against it,” said Jorge Gea, CTO, Bacula Systems. “Our new BsnapDiff capabilities represent another significant step in Bacula’s ongoing strategy of delivering deep integration with the world’s leading high-performance file systems. By eliminating many of the inefficiencies associated with traditional incremental backup methods, organizations can improve backup performance, reduce infrastructure load, and better align data protection with modern HPC and large-scale storage architectures.“

Key Benefits of Bacula’s BsnapDiff Technologies:

• Dramatically Faster Incremental Backups – Native file system tools rapidly identify changed files without requiring full scans of entire file systems
• Reduced Impact on Primary Storage – Eliminating large-scale tree scans significantly lowers backup-related overhead on production environments
• Improved Metadata Efficiency – Reduced directory reads and file status operations help minimize metadata cache pollution and unnecessary file system activity
• Shorter Backup Windows – Faster change detection enables more backup jobs to run concurrently, helping organizations keep pace with rapidly growing datasets
• Improved User and Application Performance – Reduced backup overhead helps maintain responsiveness for users, applications, AI workloads, and research environments

The benefits become increasingly significant as file counts grow. In large-scale environments where only a small percentage of data changes between backup cycles, file system-aware change detection can reduce incremental backup processing times from many hours to minutes.

For example, Bacula’s BsnapDiff for ZFS utilizes zfs diff to identify file system changes between snapshots. In environments containing hundreds of millions of files, generating a list of changed files can take less than a minute compared with many hours required for a conventional full-tree scan.

Similarly, Bacula’s BsnapDiff for GPFS leverages IBM Storage Scale (GPFS) snapshot and backup APIs to efficiently identify changed inodes without requiring directory traversal operations.

For Lustre environments, where snapshots are often impractical at extreme scale, Bacula’s Lustre Changelog Plugin consumes and tracks Lustre changelog events, maintaining an optimized record of file system changes for highly efficient incremental backups.

Part of a Broader HPC Strategy
The introduction of BsnapDiff technologies forms part of Bacula’s broader strategy of providing deep interoperability with the storage technologies most commonly deployed in HPC, AI, research, and large-scale enterprise environments.

Beyond ZFS, GPFS, and Lustre, Bacula continues to expand its native integrations with leading storage platforms, enabling organizations to leverage file system-specific capabilities wherever possible to improve backup efficiency and reduce operational complexity.

Combined with Bacula’s HPCAccelerator technology, advanced parallel processing capabilities, broad storage interoperability, and scalable archival integration, these latest enhancements provide architects with additional tools for designing next-generation backup and archival infrastructures capable of supporting rapidly expanding data volumes.

As organizations continue to build larger AI environments, scientific research platforms, and exascale computing infrastructures, Bacula’s file system-aware data protection technologies help ensure that backup and recovery performance can scale alongside the environments they protect.