For all those who aren’t IT professionals or in someway affiliated with a large enterprise and their methods of data preservation, the debate for disk versus tape seemed to be solved years ago. However, there is a subtlety in the spelling of this particular argument that makes all the difference. Although, if after you read the phrase, disk versus tape, you thought, “DVDs laid tapes to rest years ago, besides who even has a VHS player that still works?” you’re not completely wrong even if this is a different debate than intended. For the common user, DVDs solved a number of issues including reliability, performance, and ease of use; not to mention how much time you save finding your favorite scene via a chapter selection and not by manually forwarding or rewinding a tape. But the original argument, and our intended topic, addresses magnetic tape data storage and disk drives (digital magnetic storage devices and not optical discs like DVDs) both used by IT professionals and large or small companies as data storage mediums.
To begin, what are the perceived benefits of using tapes instead of upgrading to the newer technology? The major thought that drives the use of tape is that it’s cheaper per-byte of data stored than using a disk system. Though that may be true, the inherent weaknesses of magnetic tape does not allow for a “save it once and leave it until needed” archival system. According to Gartner, “data up to one year old has a 10-15% failure rate, and the failure rate of data five or more years old is 40-45%.” They further state that 34% of companies using tape never test those backups, and of the companies that did test, 77% found failures. Referring to the vulnerability of tapes and specifically this issue of failure, Nexsan has said,
It is well understood that magnetic tape degrades over time. Temperature and humidity have a dramatic impact on shelf life. Ten degrees of temperature change can change the life of a tape by ten years or more. If an administrator stores tape in a room without a raised floor, there is a great danger temperature and humidity changes will accelerate the effects of thermal decay which, in turn, will destroy data in as little as five years…The design of the cartridge and the transport are critical to tape reliability as well. The enterprise class transports used today are in the 400,000-hour range. A well-managed cartridge (correctly controlled temperature and humidity), that is also a stagnant cartridge (i.e. a cartridge that has not been used), has a shelf life of around 20 years.
While the Library of Congress and the National Media Lab recommends,
For data having permanent value, storage areas should be kept at a constant 45 to 50° F or colder (do not store magnetic tapes below 46° F as it may cause lubrication separation from the tape binder) and 20 to 30% Relative Humidity (RH) for magnetic tapes (open reel and cassette) and 45 to 50% RH for all others. Environmental conditions must not fluctuate more that ±5° F or ±5% RH over a 24 hour period. Tape should be stored in dark areas except when being accessed, being sure to keep recordings away from UV sources (unshielded fluorescent tubes and sunlight).
The failure of tape and resulting data loss devastates small businesses. Boston Computing Network states that 30% of businesses that incur a disaster and suffer data loss go out of business within a year, while 70% of those businesses will fail within five years. A common industry solution, where firms have the available funds, is to make multiple copies so that if tapes do fail there are alternate backups. Typically, a Grandfather-Father-Son (GFS) backup rotation is implemented for this purpose. This method is a hierarchical strategy where three differing backups occur according to a schedule, but usually will run daily (son), weekly (father) and monthly (grandfather) backups. However, this redundancy comes at a massive cost to the firm due to the number of tapes required to backup the mountains of data. Nexsan estimates the following numbers,
If an organization were backing up 42TB of primary disk, they would need 788 LTO-4 tapes over the course of a year. This assumes an 80% efficiency usage for each cartridge. At $38 per cartridge, the cost is $27,563. Using GFS the cost of storing 25 copies of the data would rise to $137,812. By comparison the cost of a second 42TB array as a backup target is in the range of $45,000.
Some professionals have adopted a hybrid backup system in order to combat the vulnerability of tape. This tiered system utilizes a Disk-to-Disk-to-Tape approach that leverages the advantages of online disk storage but retains the portability and perceived longer lifespan of tapes. However, this hybrid system is less cost effective and still subject to the high failure rates of tape.
How do these firms that use a GFS backup rotation rationalize tapes instead of upgrading to disk when the cost of tapes, based on a per-byte stored basis, is the major benefit attributing to their use? Nexsan’s figures don’t include the cost to manage this system, either. Tape backup equipment requires regular maintenance and cleaning, tapes must be manually reloaded when full, labeled and transported to a secure storage location; all of these operations must be closely supervised. Therefore, the cost analysis of tapes as a backup method must also include the additional tapes needed to provide secure data retention, and the administrative hours required to manage the system. A per-byte cost comparison fails to recognize the true capital requirement. Obviously these costs prevent smaller businesses from having multiple backups and are directly correlated to Boston Computing Network’s findings, that 7 out of 10 small business that suffer data loss fail.
If downtime is the main focus of concern for those large firms that give little consideration to the cost of their backup system, then tapes are not the answer. Although the physical availability of tapes may offer some individuals a sense of security, the time required to replace lost data from backup is substantial. A high performance disk system can recover 10 terabytes in about two and a half hours. While a backup system using LTO-4 tapes requires more than four days, over 100 hours, to recover the same 10 terabytes of data. With downtime and availability in mind Nexsan made this comment, “the fundamental question to ask when looking at tape or disk is, ‘What are your recovery objectives?’ After all, it’s not really about the backup, it’s about the restore. It has to work, and it has to work on time.” The extreme time requirement for data recovery from tapes should effectively resolve any argument with regards to downtime.
In conclusion, the benefits that professionals attribute to tapes are based on an incomplete analysis of the technology and the competition. The per-byte cost for tape media is less than disk, but this comparison ignores the traditional requirement for multiple backups and the associated cost to do so. Further, the downtime resulting from the extensive restoration time of a tape backup system far outweighs the advantages of physical on site availability and offline portability. The tape media is outdated, having been designed during a time where scientists couldn’t comprehend the volume of data produced on a daily basis in 2011. The amount of data produced that needs preservation far exceeds the capabilities of tape. The only question that remains is why haven’t you made the choice for a more efficient and responsible method for backing up your data?