Before I cover the dramatic industry co-operation ushered in this month by the current vendor protagonists, let's take a quick historical stroll, because what was announced by IBM on May 21, 1952, was genuinely important.
Back in 1949, when IBM's team started to develop what was to be the first tape drive, music came on 78s, radio was based on wire recorders, and computer storage was 100% punch-card based. A thoroughly delightful writeup that IBM produced in 2002 at the half-centennial of tape ("Fifty Years of Storage Innovation: Magnetic Tape and Beyond") gives insights into the engineering innovation that went into the first tape drive. It almost didn't happen because of internal disagreements about punch cards. The motor used to generate the vacuum for the vacuum columns that were the special sauce enabling tape to be handled effectively at speed in a start-stop manner literally came from a contemporary GE domestic vacuum cleaner.
[ Read Will Tape Replace The Hard Drive? ]
The path was so circuitous, you might wonder that the tape drive happened at all. But happen it did, and it immediately ushered in dramatic change. IBM's 726, the first marketed tape drive, was 56 times faster than punch cards. One 10.5-inch reel of tape could hold the equivalent data of 35,000 punch cards.
Although the absolute numbers might look quaint today (7,500 characters per second and 100 bits per inch on the tape) the relative improvement was nothing short of amazing. The 726 was announced as a part of the IBM 701 Defense Calculator, which the Wall Street Journal pronounced was "designed to shatter the time barrier confronting technicians working on vital atomic and airplane projects".
Fast forward six decades. Tape drives have become more capacious, faster, and a lot cheaper per megabyte, gigabyte, and now terabyte. Automation was successfully added once the format of the media became the now-familiar square cartridge. More recently, the traditional role of tape as a back-up medium has been augmented by (and in some places transitioned to) a role as an archive medium of choice.
Despite a constant flow of "tape is dead" jibes and assertions from non-tape vendors, the technology has refused to die. This is no doubt due to a couple of tape's basic attributes: it is capable of storing a lot of data for a very long time, largely aside from system intrusions, and--most important of all--doing it extremely cost effectively.
There's been something of a gradual renaissance of interest in tape, both as data centers have been moving toward more data tiering, and as organizations are being commanded and regulated to store ever-increasing amounts of data indefinitely. "Big data" is not just about analytics. And the renaissance hasn't been harmed by stories such as Google's recovery of 40,000 email accounts from tape a year or so ago. That the doyen of all things cool and instant was itself using tape as the ultimate data security system was music to the ears of the tape ecosystem vendors.
And yet IT's opinion of tape has barely budged in years. IT belittles tape as something you love to hate, that is prehistoric (never mind that disk drives are only four years tape's junior, first being delivered in 1956), and that is old, dusty, unreliable, and operationally tough to use. The facts belie all these denigrations of tape but the industry has--until now in any case--seemed to be happy to be a punching bag.
Not anymore. This month's assertive step forward marks an impressive--if overdue--attempt by the tape industry to mark its territory. A 2012 Tape Market State of the Union Message was signed by BDT, Crossroads Systems, Fujifilm, HP, IBM, Imation, Iron Mountain, Oracle, Overland Storage, Quantum, Spectra Logic, and Tandberg Data. What was this group crowing about?
The three biggest complaints about tape--that data is hard to find, might then be hard to read, and the media is generally unreliable—no longer hold water. The vendors emphasize that tape has become much easier to manage and use via automated health- and data-integrity verification tools. Thanks to file systems such as LTFS, tapes can look and act like enormous USB drives. Most surprising is the revelation that the raw bit error rate of tape is orders of magnitude better than that of most disk drives.
Today's tape offers a lot of potential for real value for many users--it's still the least expensive long-term storage—but it's going to require some users to think again, and re-evaluate their long-held prejudices.
Mark Peters is a Senior Analyst at the Enterprise Strategy Group, a leading independent authority on enterprise storage, analytics, and a range of other business technology interests.
The Enterprise 2.0 Conference brings together industry thought leaders to explore the latest innovations in enterprise social software, analytics, and big data tools and technologies. Learn how your business can harness these tools to improve internal business processes and create operational efficiencies. It happens in Boston, June 18-21. Register today!?