Violin Memory has created data deduplication and compression features for its Concerto 2200 data reduction system for virtual desktop (VDI) and virtual server (VSI) set ups.

The rationale is to cut the premiums of flash storage media used by minimizing the footprint of the data.

The system 2200 comprises of two high availability (HA) appliances that allow Violin’s Flash Arrays (the 6000 and 7000 models) to cut the volume of data they have to save by compressing and deduplicating it.

These ‘inline’ deduplication and compression techniques can shrink the number of bits on a board by a factor of between six and ten to one.

The appliances are targeted at ‘workload environments’ (such as data centers) according to Violin.

The Concerto 2200 HA appliances work by deduplicating and compressing shares, share groups or files from NFSs (network file systems).

The data is then written to logical unit number (LUN) partitions on the Flash array.

The LUNs being serviced can all be on one array or spread across as many as four arrays.

The compression ratios achievable mean that the equipment is capable of reducing a data load by 672TB, which is enough to fill 2,500 persistent desktops in a VDI deployment.

This works out at a cost saving of around £1.09/Gb or £45/desktop.

Violin Memory’s chief marketing officer and SVP of alliances Eric Herzog said: “These are powerful tools to maximize storage efficiency while optimizing performance at the application level.”

“Competitors offer always on deduplication and compression, but [their] performance may suffer as a result of this approach,” Herzog said.

IDC’s flash storage research director Eric Burgener said the high cost of flash arrays makes the efficient use of this storage space critical.

“Certain application workloads benefit much more from compression, while others benefit much more from deduplication,” Burgener said.

“For maximum effectiveness, these need to be implemented in-line without unduly impacting application performance. Vendors need to offer both”.