Let's try a provocative stance and argue this: Hurricane Disk is now a category five storm heading straight for the low-lying tape coast. At all levels of the tape market disk-to-disk (D2D) alternatives are flooding in and a storm surge is confidently expected to overwhelm tape's defences.
The tape levee has already been substantially weakened by Microsoft's D2D DPM product. Now hear this: every tape automation vendor better get reconciled to a market that's going to implode, or else get into the D2D and virtual tape market. Watching the entry-and medium-level tape drive and automation markets over the next three years is going to be like watching a slow-speed train wreck. Although the tape automation vendors are in denial, especially the ones without a D2D strategy, it's no use. That train wreck is going to happen.
Why is this an arguable story? I'll start with the rate of product introductions:-
Dell is introducing DataSafe, an option for US Enterprise PCs with a RAID 1 mirroring scheme. They ship with a second hard drive and use Symantec Ghost software to automatically and continuously copy files to the second disk. It is so obvious and so much needed. At a price of $99 for the option it's also darn near as cheap as a DIY solution and ten times more convenient and easy.
It must be the first safest guess in the world that the scheme will extend to Europe and the rest of the world. And the second? It will be extended to low-end servers.
Imation has introduced a disk drive in a tape cartridge so tape libraries can actually backup to disk.
In the virtual tape market announcements are becoming frenetic. FalconStor, likely used by IBM for its virtual tape product, has released a barrage of VTL announcements recently. Here's one.
Sepaton has upgraded the scalability of its VTL products, used by HP and others, substantially.
Overland Storage has announced a tier one software OEM for Protection O/S, the software in its REO D2D products, which have recently been refreshed.
Quantum has extended its D2D range with substantial new products.
Copan has upgraded its MAID product: D2D in dense racks designed as a tape library alternative.
Nexsan has entered this market too with its SATABeast using AutoMAID technology.
At all levels of the tape automation market D2D products are the barbarians at the gate. Will customers withstand their siren sales calls? The D2D suppliers are convinced of it.
Before the D2D market is in full flood even, generation two products are coming thick and fast.
Now let's argue that customers hate tape backups, they really do.
The suppliers understand what we commentators just beginning to discern. Tape backup customers are beginning to understand that they are actually starving. They need new food because tape storage has kept them on a starvation-class backup diet for years.
Only now are they realising the countless hours wasted on running those slow, oh so dreadfully slow, tape backups, those interminable verification cycles, the management time, the money wasted, the difficulty and inconvenience of restore operations.
They are beginning to comprehend that: "Up with this we do not have to put."
Disk is becoming the first backup resort, tape the last.
Tape brother, ask not for whom the D2D bell tolls, it tolls for thee.
That's the arguable case for D2D to simply wash over the tape backup market and erase parts of it for ever. Tape's glorious city never will be rebuilt where it once stood. Instead a rump settlement, much much smaller, will be built on higher ground and there the tape library temples will exist with their priests regularly circling the walls in chanting processions, singing out over and over again: "Our media is cheaper. Our media is cheaper."
But you can hear the sound of the onward wind as it picks up speed: "For how long? For how long? For how long....
[Tomorrow] But you can't take disks off site!