What is the name of the backup technology that backs up only the changes that have been made to a file rather than backing up the entire file again?

What is the name of the backup technology that backs up only the changes that have been made to a file rather than backing up the entire file again?

For example, an incremental or differential backup would completely re-backup a file that had changed since the last full backup.

However, I have heard that there is a new more advanced technology than incremental or differential backups that will backup only the changes that have been made to the file that had previously been backed up rather than re-backing up the entire file.

Therefore whenever a restore is performed on a file that had been modified since the last full backup, the contents of the original full backup will be restored and then the restored file will be modified with the contents of the subsequent backups that contain the changes and modifications to the file.

Using this sort of backup technology is much more efficient and also results in much less overall space being used for storing full backups and the subsequent later backups.

What is this sort of backup technology called?
IT GuyNetwork EngineerAsked:
Who is Participating?
@meyersd he is definitely not referring to de-duplication. He is referring, as many people have said, to block level backups. This may or may not be available in his backup software.

De-duplication is different that is a process where the backup system looks for blocks of data that are identical. It then only saves 1 copy of the identical block and keeps track of which blocks were duplicates.

Using data de-duplication means that you can indeed reduce the storage you need to backup data as each identical block is stored only once. For example when doing a VMware block level backup to my TSM server my de-dup process finds massive numbers of duplicate 32k blocks as its backing up VMware server files and since a huge number of the servers have identical OS software etc in blocks the TSM backup software can massively reduce the amount of storage needed to safely backup the server snapshots.

For a small installation like mine where we have just 23 servers I need just 3.761 Tb of data to store 12.484 Tb of server images a 74% saving (according to TSM).

The principal use of Block Level backup is instead to minimise the amount of data needed to be transferred, typically this is used in things like backing up client PCs (such as laptops) over slow network links, eg: hotel wifi or mobile "broadband" links etc. Where the less data you have to transfer the better.

The downside of block level backup is significantly longer restore times, depending on your backup strategy of course. You just need to work out how frequently you want to do a full backup to minimise the restore time.
block level
Incremental backups will only backup changes since the last full or differential backup. This usually occurs at the file level, backing up a changed file. However, some software, especially on arrays, can backup changes at the block level. In any case its an incremental backup.

Hope this helps.

Has Powershell sent you back into the Stone Age?

If managing Active Directory using Windows Powershell® is making you feel like you stepped back in time, you are not alone.  For nearly 20 years, AD admins around the world have used one tool for day-to-day AD management: Hyena. Discover why.

Block level has been around for many years in the form of rsync

Novell had a commercial product called ifolder that used that technology back in in early 2000's.
CrashPlan also implements block-level differential backups which copy only the specific parts of the file that have changed rather than re-doing the entire thing. It's very useful for things like large Outlook PST files. It also does "data deduplication" so if the same block of data appears in several different files it isn't repeatedly backed up.
Duncan MeyersCommented:
I believe you're referring to de-duplication. It backs up only the parts of a file that changes.

Backup Products that include this technology included EMC Avamar, Symantec Netbackup deduplication and CommVault

I have seen de-duplication ratios over 99.9% regularly with Avamar That means for every Terabyte you backup, you only store 0.1%. That's 1GB. Pretty impressive stuff.
We're talking about backing up delta changes and you can name it whatever,but it's been around for many years as rsync.

Duncan MeyersCommented:
rsync is a replication tool.
Its called incremental backup

If i would explain using Acronis
It creates on complete backup and after that it creates the incremental backup files of few hundred mb

easy to deploy
less space consumed
fast process
less load on network
easy to maintain

I would define a schedule as a complete backup monthly
and incremental daily or weekly
Duncan MeyersCommented:
The question is an almost text-book definition of a source-based deduplication solution.

Don't forget that you can deduplicate at the target (like your TSM solution or Data Domain or Quantum DXI or HP Storewise etc) or you can deduplicate at source where, after the initial backup, you *send and store only the changed blocks* to the backup servers. Examples of source-based dedupe products are EMC Avamar, Symantec Netbackup and CommVault.

A source-based dedupe solution has extremely fast backups because you massively reduce the amount of data that has to go across the wire. Restores are also quick but obviously slower than a backup as you send the reinflated (or rebuilt) file back
To the restore location.

Extended restore times for a full restore are a particular issue with TSM. It's always been the Achilles Heel of the product.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.