2 min read

The hard sell on hard disks

Shopping for hard disks has become a whole lot more interesting this year with the announcements, most recently by Seagate, of one terabyte drives.

These disks are proof that data expands to fill the space available. And then it expands to overflow that space, generating the need for yet more disks.At the same time, space concerns in data centres and the cost of power and cooling are encouraging IT managers to try to consolidate storage. And one way they can do so is by popping smaller drives out of their arrays and replacing them with the terabyte monsters.

Of course, that leads to other challenges. How, for example, do you back these beasts up during an ever-shrinking backup window? How do you retrieve the right file from that backup when required? How do you keep track of the data on the disks? And how do you convince users that corporate storage is no place to keep their collection of Grateful Dead music (it is, after all, too big to keep on their PC hard drive)?

While resellers can’t do much about peoples’ taste in music, answers to the other questions – and, indeed, reminding the customer that they should be asked – can be lucrative.

For example, how do you back up a terabyte of data? It would take several tapes – even LTO 4 cartridges won’t inhale a full terabyte without compression – and probably more time than is available in the average backup window. So tape is probably out as a primary backup medium. And that leaves – what? As LaCie once cheerfully told me, when I asked the same question about its portable two terabyte Big Disk, you back it up on another Big Disk. So the opportunity is there to sell twice the number of drives, simply for backup.

Hardware margins being what they are that’s probably nice, but not too exciting. However, there’s software and services revenue to be had as well. One hot area is data de-duplication.

As every administrator knows, a huge chunk of disk space is typically eaten up by multiple copies of the same files. They’re stored by different users at different times, and all concerned are oblivious to the fact that the file exists already, or figure that it’s quicker to store an extra copy (after all, disk is cheap, right) than it is to find the original.

In an effort to stop wasting time and media backing up the same corporate database (or the same set of jokes) six times, administrators are using technology, increasingly included in backup software, to determine if a file has already been backed up and, if so, not wasting the time backing it up again. Instead, all that hits the backup media for subsequent instances is a pointer to a single copy of the file. Think how much time and space that would save while backing up the average e-mail system, with its dozens of cc and bcc messages!

Couple technology like this with an offering of backup as a service and the margins can start to look much nicer.