Optimized Backups Using Extent-Based Technology

Optimized backups protect very large files with greater resilience than regular backups and are used by default for files that are 1 GB or larger for Windows servers, Windows Laptops, and Linux laptops, and 100 GB for UNIX servers. The optimized backup operation breaks large files into small, individual extents that are backed up using multiple streams (for example, a 1 GB file is broken into 256 extents that are 4 MB each, and then the 256 extents are backed up independently).

Optimized backups are useful when you back up:

  • Data that has large variations in file size.

  • Data that contains large database files on file servers.

  • Data from sources, such as laptops, that have frequent disruptions.

To enable optimized backups on subclients created prior to Feature Release 11.21, see Enabling Optimized Backups On Older Subclients.

Benefits

  • Faster backups

    Optimized backups ensure better utilization of all backup streams, so operations complete faster. With a traditional backup operation, when there is a high degree of variation in the size of the files that are backed up, some backup streams remain idle. Large files are backed up using multiple streams that increases the speed of backups.

    For example, to back up 2 large database files with 4 data streams, a traditional backup operation uses only 2 streams (1 stream per file). Optimized backups use all 4 data streams to back up the extents of the files.

    Caution

    We recommend that you use an optimum number of data streams. If you specify a high value for data streams, you might experience operating system issues such as file system cache exhaustion and hardware issues such as disk thrashing thereby affecting the performance benefits.

  • Efficient Backups

    If an interruption occurs during a traditional backup of a large file, the file is backed up again when you restart the job. With optimized backups, minimal data is retransferred, so the backup operation is more resilient and efficient.

Considerations

  • Extents-based backup is supported only on client computers that use Indexing Version 2.

  • The default file size for extent-based backups on UNIX servers is 100 GB. The default file size for extent-based backups on Windows servers, Windows Laptops, and Linux laptops is 1 GB.

  • When you use the Restore By Jobs feature, files that are backed up as extents are not restored.

  • Files that are backed up as extents cannot be restored using cloud object storage agents such as Azure File Storage and Azure Blob Storage.

  • You cannot use extent-based backups for on-demand backups and ad-hoc backups.

  • For Windows File System, optimized backup is not supported for encrypted files.

  • For UNIX File System, review the following considerations:

    • Extent-based backups are not enabled by default on Linux clients configured to back up the hard links.

    • Extent-based backups are supported only on Linux and AIX platforms. Platforms such as HP-UX or Solaris do not support extent-based backups.

    • Extent-based backups are used during backup copy jobs, but only if you enable the Skip Catalog Phase for Snap Backup check box on the Advanced Backup Options dialog box. For more information, see Skipping the Cataloging Phase for IntelliSnap Backups.

Supported Agents

The following agents support extent-based backups:

  • Windows File System

  • UNIX/Linux File Systems

  • IntelliSnap for Windows File System

  • IntelliSnap for UNIX File System

  • CIFS or NFS Backups for the following clients:

    • Generic NAS clients

    • NetApp streaming backups with SnapDiff

    • Dell EMC Isilon with Changelist API

    • Nutanix Files

    • Qumulo File Storage

  • Hadoop (HDFS)

  • IntelliSnap for SAP HANA

Loading...