Veeam Home | Support | Downloads
Best Practices for Veeam Backup & Replication 7.0 for VMware

Veeam Management Pack Help Center  > Best Practices for Deployment & Configuration >  Tape Integration > Planning for Tape Archiving > Understanding Media Sets

Understanding Media Sets

Previous page Next page Print this Topic

Table of contents

A media set is a consequent data stream that can span several tapes (for example, a weekly database backup). It refers to physical media (not to managed logical entities) and thus it is not displayed as a managed object in Veeam backup console. However, users can specify how this data stream should be organized — by using the Media Set page of the Media Pool Wizard:

tape_media_set_retention_zoom90

Naming

If you choose to create a new media set periodically (daily, or every Monday, or alike), the creation date will be appended to the default name “Media set”, so the full default name will be like Media set 11/11/2013 12:00 PM. You can use the Media Pool wizard and modify media set name to include, for example, backup type or application name: Infrastructure Full Backups 12/12/2013, Full Weekly SQL backup, and so on.

Forcing Media Set Creation

On the Media Set page of the Media Pool wizard, there are also several options for new media set creation. Consider the recommendations below when making your choice.

Option

Pros

Cons

Comment

Do not create, always continue using current media set

No wasted space on tapes.

Backup sessions can be dispersed between several tapes, and you have to search through tape archive to find the required backup file. This complicates the restore process

This option can be used if you have a limited number of tapes, and also if there are no requirements for granular retention.

Create new media set for every backup session

Support for granular retention, so you can easily locate the backup file you need.

Space can be over-allocated (for example, a backup session actually records 10GB to a 1TB tape, and the next session starts with a new tape).

If multiple backup jobs are linked to a single tape job, a backup session starts with the first backup job start, and ends with the last backup job end. In this case, it is recommended to schedule tape job so that it starts a new media set when the first linked job starts. This option allows you to segregate data from different tape jobs, i.e. keep archives created by each tape job separately.

Daily at: every [day]

You always know in which media set your backup sessions are; space is used efficiently enough.

Though retention is granular enough, you will have to look through tapes in the media set to locate the required backup (however, no need to look through all existing tape records).

This option is the best choice if you plan for daily/weekly/monthly archiving cycles with periodic tape export on the certain days. For example, if you have a backup job that generates daily increment, then it is reasonable to record the whole chain of increments to the tape once a week on Saturday, and schedule export on Sunday.

Restoring

Consider that when restoring, you cannot start the process from a random tape within a set.

As a work-around, you can run Catalog job, or run the Inventory job for the last tape in the set (as it contains metadata about all tapes included in this set).

Table of contents