NetApp debuts all-file deduplication

Network Appliance on Monday introduced deduplication for all types of file data rather than just backup data.

In what might be the first such offering in the industry, Network Appliance's A-SIS (Advanced-Single Instance Storage) supports the deduplication of data on its Fabric Attached Storage System (FAS) and its compliance and archiving-oriented NearStore appliance.

Using deduplication, customers can reduce the amount of storage they need to buy and subsequently manage. The technology had previously been used only on virtual tape library appliances and software from Diligent, Sepaton, FalconStor and Data Domain.

Davinder Gupta, manager of network systems for Intuitive Surgical, beta-tested A-SIS on a NetApp NearStore.

"We did a pilot after the beta product," Gupta says. "We took 100GB of data, which was representative of our data set -- Word files, Excel, PDFs, movies, everything -- and we turned the process on. We were seeing a 30 percent reduction in data."

A-SIS works with all types of unstructured data -- Network File System and Common Internet File System files, Excel spreadsheets, Word documents and PDF files.

The form of deduplication Network Appliance uses has been integrated with NetApp's Data ONTAP 7.2 operating system and its WAFL file system. WAFL takes only changed blocks of data and saves then to disk, thus providing single-instance storage. WAFL creates a piece of metadata, which serves as a digital signature to compare to other signatures on the volume. The company claims the deduplication process, which is done post-processing rather than inline, only causes an approximate 1 percent hit on disk writes.

A-SIS is limited to use on only Network Appliance flexible volumes and to volumes 4TB in size.

The company is working toward making A-SIS work with transaction based database data in a future release.

NetApp's A-SIS is included on NearStore and FAS appliances at no charge.