Scientific advancements continue to drive tape's market leadership.

While electronic data generation continues to grow at an 80% annual rate globally, according to IDC, the tape industry is keeping pace with the introduction of larger capacity, faster data transfers, and more scalable products that provide even better returns on investment that provides the protection of your business' data. Unlike their magnetic disk cousins (where the understood physical limits of the medium are continually being challenged and redefined) tape manufacturers have proven that there is at least another 16-fold capacity increase possible by using technologies and materials that are already being tested in their labs.

It is well established that tape is the most cost-effective media for storing large amounts of data. And even as disk capacities have dramatically increased over the past 10 years, the tape industry has continually maintained a capacity advantage over disk; a single tape cartridge can be used to backup the contents of a single hard drive. This has always been the case, and it continues to be the driving imperative behind tape drive and media development.

[GRAPHIC OMITTED]

The science behind the improvements in tape technology is very impressive, and also very technically sophisticated. Over 40 researchers representing the major tape drive manufacturers, tape media suppliers and many of the leading research universities, collaborated (as part of the Information Storage Industry Consortium--INSIC) to analyze enterprise tape market requirements. The result was the development in 2001 of a 10-year projection for the future of magnetic tape storage. INSIC's work indicates that to remain an economically viable storage solution when compared to disk, tape capacity must grow at a rate comparable to future disk capacity growth on a cost-per-gigabyte basis.

Although disk densities are continuing to rise, the rate of growth is expected to slow as disk recording nears a superparamagnetic limitation. Due to a much lower areal recording density and much greater recording area, tape technology has the potential to grow at a faster rate and as a result improve its cost per gigabyte trends compared to disk. Therefore, when combined with disk in the enterprise storage environment, the tape industry roadmap currently maintains that tape capacity on a single cartridge must achieve 10 terabytes (TB) uncompressed per cartridge by 2011, and must reach 1TB uncompressed on a single cartridge by 2006 on its way to reaching the 10-year goal.

Areal Density 

One way to increase storage capacity is to increase the areal density of the medium. Areal density is the number of bits that can be reliably stored on, and retrieved from, a fixed surface size, such as a square inch on the tape or disk. In the mid-1990s, physicists announced the theoretical areal density limit of magnetic disk at 36GB/[in.sup.2], and expected to hit the actual limits somewhere before this theoretical limit. Today, that limit has been surpassed and driven to top-of-the-line enterprise disk drive models of 75GB/[in.sup.2].

Today's mid-range quarter-inch linear tape drives can store data at about .25 to .35 GB/[in.sup.2] on tape media, which is 200 times less than that of a mid-range magnetic hard disk of 65 GB/[in.sup.2] on a 3.5-inch platter. If you consider a half-inch linear tape cartridge, it is 1000 times that of a 3.5-inch platter. The capacity advantage of tape is its large volumetric density of tape in a given cartridge. While this may seem like tape is technically behind disk, tape drive design has kept pace with overall capacity increases, while tape reliability and data retention longevity have increased.

"Disk and tape scientific and engineering advances have been applied to each technology's products," states John Goode, Quantum's Tape Storage Product marketing manager. "These include the development of new, smaller magnetic particles, improved methods of placing these particles on the medium, and smaller, more responsive read/write heads with more parallel data channels."

Greatly improved methods for detecting and recovering from errors have also helped to push the limits of tape and disk capacity. Simple parity schemes have been replaced by complex algorithms, including Partial Response Maximum Likelihood (PRML), to make intelligent decisions on what the recorded content really should be.

As improvements in magnetic media design continue, the developers of tape drives and media have more technology to work with to improve their products. But even if and when magnetic media (particularly disk engineers) do hit the areal density wall, tape engineers will still have many generations of capacity improvements to come, due to the 800:1 ratio in areal density that currently exists between disk and tape design. Another way to look at this would be to say that current tape cartridges contain 800 times the recording surface of a disk drive.

For example, suppose the scientists working on disk drive design said today that they had reached the physical capacity limits of hard disks, at 300GB for a 3-inch drive with 4 platters. That would put the theoretical capacity limit of a mid-range tape cartridge, using recording technology available today, at over 20TB. Tape manufacturers are claiming far more conservative goals today, but their scientific research is all about pushing the limits to provide even greater value to their customers in the future.

Tape Length

Since the size and number of disk platters are fixed in each defined disk drive format, areal density is really the only parameter that can be improved to increase storage capacity. In fact, as the bits on the disk get closer together, a greater percentage of the disk needs to be reserved for error correction (ECC) operations. Tape, however, does have one major capacity-increasing advantage over disk: the length of the tape can be increased to provide a larger recording surface. In a fixed-sized cartridge format, the tape can be lengthened in conjunction with using thinner materials to manufacture the media, so that more tape can fit in the same cartridge space. With thinner/longer tape, the tape path mechanics of the drive are also re-designed to safely move the thinner tape at high speeds, since the market expects increased performance with the increases in capacity. Slowing down the transport of the tape over the tape head would allow for much greater capacity, but this is not an acceptable option in most cases.

Intuitively, one would think that a thinner tape wouldn't have any impact on the recording electronics.

However, the depth of the magnetic layer does affect the strength of the recorded signal, requiring more sensitive read heads and a higher tolerance to noise levels. Today's Magneto-Resistive (MR) heads meet this challenge, while new break-throughs in head technology are in the works.

"In 1991, IBM's work on AMR technology led to the development of MR (magnetoresistive) heads capable of the areal densities required to sustain the disk drive industry's continued growth in capacity and performance," says Paul Scheuer, IBM's strategy manager, Tape Storage. "This technology circumvented the fundamental limitation of TFI heads--the fact that their recording alternately performed a conflicting task-writing data as well as retrieving previously written data--by adopting a design in which read/write elements were separate, allowing each to be optimized for its specific function."

The various improvements offered by MR technology amount to an ability to read from areal densities about four times denser than TFI heads at higher flying heights. In practice, this means that the technology is capable of supporting areal densities of at least 3Gbits/[in.sup.2]. The technology's sensitivity limitations stem from the fact that the degree of change in resistance in an MR head's magnetic film is itself limited. Over the past few years, a logical progression from MR technology called Giant Magneto-Resistive (GMR) technology has further increased the areal densities to 10Gbits/[in.sup.2] and more.

Tracking

Linear tape drives such as SDLT, LTO Ultrium, and 9840, where tracks are written horizontally across the entire length of the media, have employed pre-written servo tracks on the tape itself to help keep the tape drive heads in line with the tape. As the tape goes past the read head, a special servo pick-up head reads the signal level of the servo track, and causes microscopic adjustments to the vertical position of the head assembly on the fly to keep it on track.

In contrast, helical scan tape formats such as AIT, SAIT and VXA write data in much shorter tracks across the width of the tape, and therefore do not require servo tracking systems. The structure of the disk tracking system is contained in the File Allocation Table which keeps data sectors allocated in groups called clusters. The number of sectors in a cluster depends on the cluster size, which in turn depends on the partition size.

When the computer wants to read data, the operating system works out where the data is on the disk. To do this, it first reads the FAT (File Allocation Table) at the beginning of the partition. This tells the operating system in which sector on which track to find the data. With this information, the head can then read the requested data. The disk controller controls the drive's servo-motors and translates the fluctuating voltages from the head into digital data for the CPU. Both the linear and helical scan tape drives and the disk drive keep an accurate record of the location of data on the recorded media.

Summary

While the specific technologies vary among tape formats and manufacturers, tape drive and media engineers across the industry are hard at work developing the most cost-effective products possible to meet the growing data storage challenges of the future. Tape will continue to be the best place to reliably store backup and archive data for the remainder of this decade and beyond. The current and future of tape is indeed very strong. Industry roadmap goals for tape capacities of 1TB by 2006 and 10TB by 2011 set up a future vision. The combination of capacity, performance, and reliability will meet customer needs and industry acceptance. The 50-year innovation of tape manufacturers and its leadership in associations, such as INSIC and the Tape Technology Council, shows the strength of tape recording technology for years to come.

www.tapecouncil.org

Rich Harada is president of the Tape Technology Council, (Hillsdale, NJ)