What is BackupExec Deduplication

Suppose we have BE2012 environment. I need to understand Deduplication.

Suppose we have two clients. Each clients have some same files exist. Both clients have different retention set for their backups. Suppose client A backed up and after some time client B. But client B files will not be backed up(which are same between A and B) on to BE media server as they already exist on BE media server because of Deduplication feature. What happen when Client A backup expire ? How client B restore the files as client B files were not backed up due to these files were already backed up from Client A data.

 

1 Solution

Accepted Solutions
Accepted Solution!

Backup A runs During the

Backup A runs

During the backup every file is split into chunks and the chunks are then compared with existing chunks in the Dedup Store and if any do not exist these chunks are copied to the store and the dedup database is updated to list which chunks belong to Backup A

If Backup A is the first every backup of course every chunk is copied  and 1 record to state linked to backup A is made in the database for every chunk

Backup B runs and again the data is split into chunks with the same comparisons being done as per Backup A. Any chunks that do not exists are copied and the DB records amended. Any Chunks that already exist don't copy the chunk again but still add a record to the database to show that the chunk is related to both Backup A and Backup B

When the backup set expires ONLY the database records are amended/removed for theat specific backup set, any records for other sets on the same chunk and the actual chunks remain in place.

Then at a future point in time a background process runs to see if any chunks exist that have 0 associated records for backup sets, if they show up with 0 they are removed.

This is a simple version of everything that takes places as obviously the BE catalog files and the BEDB content also experiences tasks trelated to expiry.

 

 

2 Replies
Accepted Solution!

Backup A runs During the

Backup A runs

During the backup every file is split into chunks and the chunks are then compared with existing chunks in the Dedup Store and if any do not exist these chunks are copied to the store and the dedup database is updated to list which chunks belong to Backup A

If Backup A is the first every backup of course every chunk is copied  and 1 record to state linked to backup A is made in the database for every chunk

Backup B runs and again the data is split into chunks with the same comparisons being done as per Backup A. Any chunks that do not exists are copied and the DB records amended. Any Chunks that already exist don't copy the chunk again but still add a record to the database to show that the chunk is related to both Backup A and Backup B

When the backup set expires ONLY the database records are amended/removed for theat specific backup set, any records for other sets on the same chunk and the actual chunks remain in place.

Then at a future point in time a background process runs to see if any chunks exist that have 0 associated records for backup sets, if they show up with 0 they are removed.

This is a simple version of everything that takes places as obviously the BE catalog files and the BEDB content also experiences tasks trelated to expiry.

 

 

Excellect. Thanks for your

Excellect. Thanks for your kind words. Marking your post as solution.