LTO-5 tape comes from Quantum and Overland Both Quantum and Overland have announced LTO5 tape products, but the LTO consortium still has nowhere for tape users to go after LTO-6. Linear Tape Open (LTO) 5 is the fifth LTO format and specifies cartridges holding 1.5TB of raw data with a 140MB/sec transfer rate. SpectraLogic was first to announce systems, in September …
" The lack of a roadmap might incline customers to look more favourably upon disk-to-disk (D2D) backup, especially deduplicating disk-to-disk backup. We can expect D2D arrays to be employing 4TB drives, possibly even 8TB ones, when LTO-6 tape drives appear in roughly three years time. Unless the LTO Consortium gets its ass in gear and presses its tape pedal to the metal, it could find that its lack of urgency will help bring about its own demise."
Yeah right tell that to the likes of CERN. Tape arrays and libraries exist and will still exist because they are more reliable, use a lot less electricty and tapes have a certified offsite shelf life (archiving / manditory data retention) of many years unlike HDDs. Unless that D2D is mirrored offsite you can forget any insurance cover for fire, flood and theft.
As for deduplication I am not sold as any large organisation and some small ones already emloy SANS that do that as part of their job. There is little to no value deduping at backup as the bigger issue is why all that duplicated material sat on far more expensive first line storage in the first place. Even the dreaded windows servers in 2008 / 2003R2 versions have deduplication capability in the storage server versions.
First off I don't buy into the tape is more reliable than disk argument, the most unreliable facet of most backup solutions is the wear and tear associated with tape mechs and bad policies as far as media retention is concerned (i.e. people using tapes for more backups than is practicable) that are mainly down to human error.
Backup to disk is proving to be far more reliable (because of RAID) and less prone to job failure, it can also address challenges around speed of backup (though if you can stream it tape can be quicker 0 thats a big if!) and recovery from backup images can also be quicker and more reliable.
De-dupe helps purely in a B2D context as it reduces the physical number of disks you need to store multiple backup images - these savings can be huge. Many VTLs or Backup software vendors include facilities to automate the migration of older backup jobs to tape. So whats being de-duped here isn't neccessarily the same as what you might be de-duping in your primary storage i.e. on a NetApp FAS for instance. If you backup a de-duped primary volume to tape it won't stay de-duped on tape as it will re-hydrate into its full size so you get no benefit as far as backup is concerned for de-duping primary storage its more about saving cost.
That said Im not of the brigade that thinks theres no place for tape. I think the reality is that you can accomodate both, tape for longer term storage and recovery and D2D for shorter term reqirements as part of a comprehensive backup strategy to address a variety of RPO/RTO requirements
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
