Cloud Native Scale-out Deduplication: Remove Purpose Built Backup Appliances, Store to Object Storage and Save Up to 70%

by Hugh Emberson, CTO

StorReduce Scale-Out Deduplication Now Enables Primary Backups of On-Premises Environments Straight to AWS Cloud and Removal of Purpose Built Backup Appliances.

StorReduce helps enterprises storing unstructured data to Amazon Simple Storage Service (Amazon S3) or Amazon S3 Infrequent Access to speed up transfer of the data and reduce their amount and cost of storage by as much as 98 percent.

StorReduce’s deduplication engine is software defined, deploying on-premises or on-cloud as a virtual machine, in a docker container or on physical hardware. The deduplication removes any redundant blocks of data before it is stored and ensures that only one copy of each block is stored. It scales to Petabytes of storage, all with a cloud native interface (S3 API) enabling you to seamlessly use the deduplicated data on AWS Cloud with AWS services such as Amazon Elastic MapReduce (Amazon EMR) for data mining, and Amazon CloudSearch.

StorReduce enables movement of secondary or archival data off tape or on-premises NAS at massive scale to AWS Cloud. It reduces transfer bandwidth, backup windows and cloud storage costs by up to 50X with deduplication on premises with the meta data and writeable version on cloud as well.

StorReduce’s new Scale-out Deduplication offering for the first time enables enterprises to do primary backups straight to AWS Cloud and recover at LAN speed when they have a Direct Connect, thus enabling them to remove on-premises purpose built backup appliances. The benefits to the customer are:

  • Savings of 70% or more for your entire primary backup environment
  • Escape the hardware cycle of re-purchasing all software and hardware typically every 3 years for their on-premises backup appliance
  • Scale out storage as you need it on commodity hardware, with the scalability of the cloud
  • The data now enjoys the durability and redundancy of AWS Cloud which an on-premises appliance does not have
  • Feature enablement on that data on AWS Cloud that you would not otherwise have, using generic cloud services via StorReduce’s open, scalable, S3 API interface

StorReduce Scale-out Deduplication works like this:

The key characteristics of StorReduce Scale-out deduplication are:

  • Scaling factor of over 95%
  • One StorReduce cluster is estimated to match the throughputs of 5 of the largest purpose built backup appliances
  • StorReduce does not charge any additional fees for additional instances or for multiple sites, making it scalable and ideal for the largest enterprises and Petabyte scale deployments
  • Sustained inline deduplication 247 with no buffering
  • Stateless: All data is confirmed as having been received by StorReduce only after it is actually written to the cloud
  • Software defined: can deploy in around 30 minutes
  • Speeds up interactions with S3:
    • Data transfer to S3 is deduplicated by up to 50x and compressed
    • StorReduce is highly parallel. One transaction with StorReduce might be split into 10 simultaneous transactions with S3 - much faster.

This is what a typical legacy primary backup environment looks like currently:

Here is an example architecture of primary backup straight to Amazon S3 or S3-IA:

One Scalable, Cloud Native Deduplication Across All of Your Data On-Cloud Enables a Cloud First Future

A typical enterprise will have a mix of backup applications, some of which may have built in deduplication. You can’t architect a cloud first future for your company by moving data into cloud with the wrong deduplication offering(s). The wrong offerings tie you into an on-premises appliance hardware cycle which is expensive, not scalable and not redundant. If you move your data to cloud via a range of differing deduplication technologies, you can’t join all of your unstructured data on-cloud in one amorphous blob (or data lake). You can’t use deduplication to benefit all of your cloud data. In addition, few of the deduplication technologies have scale-out, high availability or a cloud native interface, so your data is stored in silos on cloud and is not able to be used in future with generic cloud services such as search and data mining to extract competitive intelligence from your data.

StorReduce works in or behind all major backup appliances’ cloud connectors (like NetBackup, CommVault, Veeam, Oracle RMAN, EMC NetWorker and Arcserve) and it typically requires a mere configuration change to use StorReduce to deduplicate and send any backups, including Hadoop backups to AWS cloud. This frees your data on cloud into a single scalable pool with a cloud native interface and cloning.

To request a StorReduce Scale-out free trial, contact . To calculate your Total Cost of Ownership and savings from removing primary backup appliances and/or tape in your environment and migrating that data to cloud / object storage, click here (backup appliances, save up to 70%) or here (tape, save up to 45%) and enter values in the relevant TCO Calculators to be emailed the workable spreadsheet and assumptions.


Subscribe in a reader

Or subscribe via email: