VB365 Backup Repository Server
General backup repository design:
- One Backup Repository can only be linked to a single Backup Proxy Server.
- Disk Repository
- Object Storage Repository
- Local Cache
- Object Storage data size
- API call cost estimation
- Immutability
Important: When creating a bucket, only enable versioning in combination with Object Lock (immutability). Never enable versioning when immutability will not be used.
- Using Object Storage with Veeam Products: KB4241
- Supported Object Storages and repository structure: User Guide
- User Guide
Retention policy
VB365 has two retention policies: snapshot-based (=default) and item-level. Depending on your client’s requirements, one or the other will be chosen.
-
Use snapshot-based retention to keep all items and versions for the configured retention period. Similar to how image-level backups work.
Example: Backup all items (full backup) and then backup only the new/changes (incrementals) and keep for 3 years. All items will be kept for at least 3 years.This retention type is the default and is most used to make sure that all the items and versions are backed up and kept for the required period agreed with the client.
-
Use item-level retention to keep items based on its creation time or last modification time and that that fall within the configured time period.
Example: Backup all items with a last modification date from within the last 3 years. Items not modified within the last 3 years are not even be backed up.
Items will also be deleted from backup once their creation or last modification date exeeds the configured retention period.This retention type is used in certain compliancy scenarios.
Example: In the HR department, it might not be allowed to store items older than 3 years for legal and compliance reasons.
Therefore items created or modified longer than 3 years ago, must be removed from backup.
More details and examples: User Guide
Sizing
When it comes to Backup data disk sizing, it depends on whether you are using JetDB-backed repositories or Object Storage:
JetDB-backed repositories
Store all of the data 1:1 on disk.
The maximum size of a JetDB is 64 TB per (.adb) file. Once 59 TB is reached, a new (.adb) file is generated in the same storage location.
Veeam stores items in a JetDB per year based on the items’ creation time and last modification time.
Even though retention gets upon the scheduled interval, the database files will never shrink on their own.
Once retention is applied and backup data is removed, the space inside the database is freed up to be reused by new backups.
Size accordingly to how much source data you need to backup + retention.
- Example: 3 year backups with 10 TB per year will result in 3 JetDB files on disk and 30 TB in total.
Object Storage
Use JetDB-backed repositories as cache to avoid most API calls.
As a rule of thumb provide 1% of the source data size as local cache for metadata.
- Example: 10 TB source data will result in 100 GB as JetDB-backed local cache(s) on disk.
Use the official VB365 Capacity Calculator to estimate storage consumption.