We want to incorporate deduplication into our backup regime. What are the key factors to consider when deciding...
whether to go for target or source deduplication or a mixture of the two?
Data deduplication at the source is best suited to use at remote offices for backup to a central data centre or the cloud. It is usually a feature of the backup application. Target deduplication is best suited to use in the data centre for the reduction of very large data sets and is usually carried out with a hardware appliance.
So, when considering whether to deduplicate at source or target, the best deduplication process option of all is to go for a hybrid model that uses each capability in the most appropriate areas of your environment.
It’s likely to be an approach that suits most organisations too as nearly all have remote sites, workstations and servers that need to be backed up as well as mission-critical centralised IT resources.
On the one hand, source-side backup deduplication is an effective tool for scenarios where data can be deduplicated before it’s sent over the wire, creating a bandwidth-efficient backup.
Reducing the data transfer will also in most cases reduce the backup window, allowing your backups to complete in a shorter time frame and reducing the load on centralised resources. Deduplication at the source does more than save storage space; it has a positive impact on network and operational resources and application performance.
Source deduplication also allows IT organisations to write data efficiently to the cloud. And customers benefit from backup deduplication in cloud services as the amount of data they send to the service provider is reduced, saving them money.
Performing the deduplication process at the target, on the other hand, often requires a hardware appliance and is therefore more costly in terms of capital expenditure, but it is well suited to organisations that need to reduce data volumes significantly within the data centre and where bandwidth isn’t a consideration.
Target-side deduplication is all about rapidly reducing the footprint of large data sets, lowering the cost of ownership and sweating the existing assets.
Ultimately, source and target strategies overlap efficiently to provide benefits across the backup landscape. Instead of choosing one or the other, most organisations need to have a hybrid approach to maximise on its benefits. How this is done, however, depends entirely on the makeup of your organisation.