Data Compression and Deduplication

Veeam Backup & Replication provides mechanisms of data compression and deduplication. Data compression and deduplication let you decrease traffic going over the network and disk space required to store backup files and VM replica files.

Data Compression

Data compression decreases the size of created files but affects the duration of the backup or replication procedure. Veeam Backup & Replication allows you to select one of the following compression levels:

• None compression level is recommended if you plan to store backup files and VM replica files on storage devices supporting hardware compression and deduplication. Disabled compression can reduce performance due to the increased amount of data to be transferred.
• Dedupe-friendly compression level is recommended for some deduplicating storage appliances and when external WAN accelerators are used.
• Optimal compression level is the recommended compression level. It provides the best ratio between compression and performance, the lowest backup proxy CPU usage and the fastest restore.
• High compression level provides up to 60% of additional data reduction over the Optimal compression level at the cost of 2x higher CPU usage and 2x slower restore.
• Extreme compression level provides up to 33% of additional data reduction over the High compression level at the cost of 5x higher CPU usage.
• [For backup copy job] Auto is the recommended compression level for backup copy jobs. Select this level to use the compression settings of the copied backup files.

Changing Data Compression Settings

You can change data compression settings for existing jobs. After you change the settings, you will not need to create new full backups to use the new settings. Veeam Backup & Replication will automatically apply the new compression level to the newly created backup files after you save the settings. Previously created backup files will not be affected.

However, if you use the reverse incremental backup method, the newly created backup files will contain a mixture of data blocks compressed at different levels. For example, you have a backup job that uses the reverse incremental backup method and the Optimal level of compression. After several job sessions, you change the compression level to High. In the reverse incremental backup chains, the full backup file is rebuilt with every job session to include new data blocks. As a result, the full backup file will contain a mixture of data blocks: data blocks compressed at the Optimal level and data blocks compressed at the High level. The same behavior applies to synthetic full backups: synthetic full backups created after the compression level change will contain a mixture of data blocks compressed at different levels.

If you want the newly created backup file to contain data blocks compressed at one level, you can create an active full backup. Veeam Backup & Replication will retrieve data for the whole VM image from the production infrastructure and compress it at the new compression level. All subsequent backup files in the backup chain will also use the new compression level.

Deduplication

Data deduplication decreases the size of files. With data deduplication enabled, Veeam Backup & Replication does not store the resulting file identical data blocks and space that have been pre-allocated but not used.

We recommend enabling data deduplication if your backup jobs contain several VMs with a lot of free space on their logical disks or VMs with similar data blocks — for example, VMs created from the same template. However, note that data deduplication may decrease job performance.

Veeam Backup & Replication uses Veeam Data Movers to deduplicate VM data:

• Veeam Data Mover in the source side deduplicates VM data at the level of VM disks. Before the source-side Veeam Data Mover starts processing a VM disk, it obtains digests for the previous restore point in the backup chain from Veeam Data Mover in the target side. The source-side Veeam Data Mover consolidates this information with CBT information from the hypervisor and filters VM disk data based on it. If some data block exists in the previous restore point for this VM, the source-side Veeam Data Mover does not transport this data block to the target. In addition, in the case of thin disks, the source-side Veeam Data Mover skips unallocated space.
• Veeam Data Mover in the target side deduplicates VM data at the level of the backup file. It processes data for all VM disks of all VMs in the job. The target-side Veeam Data Mover uses digests to detect identical data blocks in transported data and stores only unique data blocks in the resulting backup file.

You can change data deduplication settings for existing jobs. After you change the settings, you will not need to create new full backups to enable or disable the deduplication. Veeam Backup & Replication will automatically apply the change to the newly created backup files after you save the settings. Previously created backup files will not be affected.

Storage Optimization

To optimize job performance and storage usage, Veeam Backup & Replication allows you to choose the minimum data block size to process VMs. The optimal data block size depends on the type of storage you select as a target and the size of your files.

When selecting the data block size, consider the following aspects:

• When reading the VM image, Veeam Backup & Replication "splits" the VM image into blocks of the selected size. The more data blocks there are, the more time is required to process the VM image.
• [For replication and VMware Cloud Director replication] Veeam Backup & Replication writes information about every data block to the VM replica metadata stored in the backup repository. The more data blocks there are, the more metadata is written to the backup repository.
• [For changed block tracking enabled] During incremental job runs, Veeam Backup & Replication uses CBT to define changed data blocks in the VM. The larger the size of the found changed data block, the greater the amount of data transferred to the target site.

The incorrectly chosen data block size may decrease the performance. For example, when you deduplicate a large backup file to small data blocks, Veeam Backup & Replication produces a very large deduplication metadata table, which can potentially overgrow the memory and CPU resources of your backup repository. For large backup files, it is better to use large data blocks.

Veeam Backup & Replication provides several storage optimization options with different block sizes used. The following table will help you to choose the optimal option according to the size of your backup files and the storage type.

Changing Storage Optimization Settings

You can change storage optimization settings for existing jobs. New settings will not have any effect on previously created files in the chain. They will be applied to new files created after the settings were changed.

For Veeam Backup & Replication to apply the new settings, use the following instructions.

Backup Jobs

To apply new storage optimization settings in backup jobs, you must create an active full backup after you change storage optimization settings. Veeam Backup & Replication will use the new block size for the active full backup and subsequent backup files in the backup chain.

Backup Copy Jobs

To change data block size for backup copy jobs, you must perform the following actions:

1. Change the data block size in the initial backup job settings.
2. Create an active full backup with the initial backup job.
3. Create an active full backup with the backup copy job.