NetApp last week officially confirmed its fait accompli to discontinue development efforts for the NetApp VTL (Virtual Tape Library). While not surprising, it does highlight a few ongoing changes in the disk-to-disk (D2D) backup world.VTL was hot 5+ years ago; virtual tape could improve your backup performance, create physical tape faster, etc. Integration was also a breeze, given how it walked and talked just like real tape. It was brute-force, hardware-compressed, straight backup -- with ingest rates of 8+ TB/hour.
But industry evolution ("feature-creep") towards virtual tape deduplication, replication, and so forth quickly become the norm.
Now fast-forward to 2010: Lack of market demand for *pure* fiber-channel connected virtual tape systems is issue #1. These days, every single D2D backup vendor supports Ethernet-attached (IP) disk targets.
Don’t believe me? Even Frank Slootman (of EMC) recently resonated the decline of the traditional tape augmentation market stating, "People think Data Domain is a VTL, but 90% of the systems we sell are IP-connected, not with Fibre Channel protocol."
In other words, “How relevant are fiber-channel VTLs moving forward?” From my conversations with partners, new deployments of fiber-channel VTLs are not top-of-mind for the vast majority of IT managers.
This realignment also confirms backup trends seen within the last 18 months: data protection around virtual machines and virtual storage isn't built (or optimized) with legacy fiber-channel VTLs in-mind. Good news: this is where NetApp products such as SMVI, Data Motion, SnapVault/OSSV, etc. really shine. And by natively deduplicating data at the source, NetApp creates end-to-end efficiency -- thus eliminating the need to dedupe archival backups.
Agree? Disagree?? Let me know!
