There are several efficiencies SimpliVity’s hyperconverged infrastructure customers gain by deduplicating all data once and forever: reduced CPU consumption, reduced disk IOPS, reduced capacity consumption, improved utilization of cache, faster backups and restores and optimized data mobility across the WAN.
Data Mobility over the WAN Is More Efficient When Deduplicated
Deduplicating all data once and forever includes data moved to remote data centers. One of our solution architects had a great blog using a Lego analogy to describe how we intelligently execute remote backups between data centers. This opens up a world of geographic data movement that has never existed before. Imagine the possibility of being able to move a complete VM from Florida to Santiago, Chile.
In fact, we have a partner that has created a lab environment to do just this. Separated by 4200 miles and utilizing an 8Mbps WAN link with 150ms latency, they were able to complete a remote backup of a 450GB VM in 76 seconds. After a 56 second VM restoration process, they began the power-on operation of the restored VM.
This operation is possible because of the globally-aware deduplication engine that is the basis of SimpliVity’s Data Virtualization Platform. SimpliVity doesn’t use traditional array-based replication concepts. Instead, the receiving node determines what data needs to be moved across the WAN, rather than the originating node forcing everything it detects as a change across.
Another big advantage of our approach to data mobility is the initial transfer of data. Often times, two storage arrays are staged next to each other to complete initial sync quickly without impact to the WAN. This brings risks due to the need to physically move the array after that initial sync, not to mention the extra effort to change network configurations and re-establish the link. One of our customers was unable to complete remote backups in 24 hours with their legacy infrastructure, but after implementing SimpliVity hyperconverged infrastructure, they were able to complete all their backups across the same WAN link with no seeding of data. With SimpliVity, only truly unique blocks need to be replicated, so syncing one Windows 2012 server also syncs the core Windows files for all the other Windows 2012 servers.
For most customers, this massive reduction of data movement across the WAN allows them to backup data more frequently to the other site. The deduplication of backups also allows them to retain more restore points. This means less lost data in the case of a disaster, which can mean the difference between the business continuing or not.
Within our hyperconverged infrastructure customer base, we’ve found that most have two locations or more. Some of them had inter-site connectivity that either had too little throughput or too high latency. This prevented them from achieving their Recovery Point Objectives (RPOs) for disaster recovery, because traditional replication simply took too long to complete. Using SimpliVity’s technology, they can now have a true disaster recovery configuration that meets their company’s needs.
Interested in learning more about the benefits of hyperconverged infrastructure? Watch our upcoming webinar with Forrester research on data protection on July 14.