Tape Storage vs. Disk Storage: Getting the Facts Straight about Total Cost of Ownership Calculations

3 minutes
3 minutes
Reading Time: 3 minutes

Modern tape storage has long been recognized for its low cost. Several analyst white papers have been published that demonstrate the low cost of storing data on tape. For example, “Quantifying the Economic Benefits of LTO-8 Technology” is a white paper that can be found on the LTO.org website. However, occasionally a storage solution provider publishes a white paper that claims to show that their solution is less expensive than tape storage for a particular use case. A good example is a recent white paper published by a disk-based backup-as-a-service provider who will remain unidentified out of respect for what they do. For the purpose of this blog, let’s call them “BaaS.” So let’s dig into their analysis which makes several assumptions that result in higher costs for tape storage than most users would experience.

Total Cost of Ownership (TCO) Process

The first step in developing a Total Cost of Ownership (TCO) estimate is the determination of the amount of data to be stored. The BaaS whitepaper separates the amount of primary data, which we wish to protect, from backup data, which is the data physically stored on the backup media. They estimate the amount of backup data residing in the tape library to be two to four times the primary data. This is due to their use of the old daily/ weekly/monthly/ full backup methodology for estimating the amount of backup data. The result is that two to four times the amount of primary data ends up being stored on tape, raising the tape hardware and media costs by two to four times.

Furthermore, for the BaaS solution, the ratio of backup to primary data is .5X to 2X. They claim they are incremental forever, and in-line compression reduces the amount of backup data required. However, today’s reality is that the incremental forever methodology is used by popular data protection software suppliers with tape storage, providing the same reduction in backup data requirements.  Also, modern data protection solutions incorporate disk storage cache, snapshots, and deduplication that all provide significant benefits and integrate with tape storage.

In addition, the capital costs for the tape solution in the BaaS white paper are significantly overstated.  BaaS estimates that the five-year capital cost of tape drives, libraries, and media servers as $150/TB to $250/TB and amortizes these costs over five years to generate an annualized capital costs of $30-$50/TB/Year. Using the Fujifilm TCO tool, we can estimate tape storage capital costs, including library, drives, and media. For one PB of backup storage, the total five-year capital cost is $35.56/TB or a little over $7 TB/Year. It is also worth noting that a tape library typically is used for longer than five years, so even the $7/TB/Year is on the high side of the tape solution capital costs.

Role of Management Costs

The final significant tape cost component overestimated in the BaaS paper is management costs. BaaS estimates tape management time of .25 hours/TB/Year resulting in a $14.42 TB/Year for tape management, assuming an IT administrator costs $120,000/year. We surveyed very large tape storage users, and the average administrator in these environments managed 42 PB of tape storage. At the same FTE cost of $120,000, our sample’s $/TB is $2.82/TB/Year for storage management, 80% less than the BaaS estimate.

With just these two adjustments the cost of tape backup storage is $34/TB/Year less expensive than the BaaS tape storage estimates. Further, if the ratio of backup data to primary data is the same and we accept the average of all the other tape cost assumptions made by Baas, we would find that the tape solution is 50% less expensive than the BaaS white paper claims. This would not be surprising as typical TCO models show as much as an 86% lower TCO for tape compared to disk.

It is also worth noting that it would be a rare organization that keeps only one copy of backup data. Most organizations require multiple copies, with one of them on tape media. Tape storage by its nature can be easily stored off-line and this provides protection against malware via a tape air gap. Tape is also very energy efficient, and with the large amounts of data that organizations must backup, this is a significant benefit. Given the importance of data security considerations and its low costs, tape will continue to be a vital component of a robust data protection architecture.

Rich Gadomski

Head of Tape Evangelism

As Head of Tape Evangelism for FUJIFILM Recording Media U.S.A., Inc., Rich is responsible for driving industry awareness and end user understanding of the purpose and value proposition of modern tape technology. Rich joined Fujifilm in 2003 as Director of Product Management, Computer Products Division, where he oversaw marketing of optical, magnetic, and flash storage products. Previously Rich held the position of Vice President of Marketing, Commercial Products, where he was responsible for the marketing of data storage products, value added services and solutions. Rich has more than 30 years of experience in the data storage industry. Before joining Fujifilm, Rich was Director of Marketing for Maxell Corp. of America where he was responsible for the marketing of data storage products. Prior to that, Rich worked for the Recording Media Products Division of Sony Electronics. Rich participates in several industry trade associations including the Active Archive Alliance, the Linear Tape-Open Consortium (LTO) and the Tape Storage Council. Rich also manages Fujifilm’s annual Global IT Executive Summit. Rich holds a BA from the University of Richmond and an MBA from Fordham University. FUJIFILM Recording Media U.S.A., Inc., is the leading manufacturer of commercial data tape products for enterprise and midrange backup and archival applications and provides long term data storage products and software through its FUJIFILM Data Management Solutions team.