IT Management

Storage & Data Management: The Future of Tape

The magnetic tape industry is changing. In recent years, tape has been relegated to the data center and the middle- to high-end of the Small to Medium Business (SMB) market with decreasing penetration below this level. Disk and tape markets have been shifting roles as disk encroaches on tape’s traditional backup/recovery market; tape is positioning itself to compete in the fast-growing tier 3 storage market that includes fixed content (data that doesn’t change), compliance, and long-life archival applications. Tape remains a viable complement to disk due to its lower cost per gigabyte, lower operating expenses, and lower energy costs. Tape provides a greener storage solution than disk drives, and is suitable for onsite and offsite storage of several generations of backup and archival data. Presently, few would predict resurgence in tape adoption or an expansion of the tape market. The notion that tape is dead ignores the substantial evidence that favors tape as a lower-cost, environmentally friendly, removable medium well-suited for high-growth compliance, fixed content, and archiving applications.

Market visibility and media coverage for tape have steadily declined in the past 10 years. The primary tape suppliers are spending less effort raising awareness or promoting the benefits of tape, and as a result, much of the IT industry has an outdated perception of today’s magnetic tape technology. Tape suppliers need to step up their marketing and educational activity as tape technology has significantly improved.

Expectations for tape include:  

  • Cartridge capacities should continue with unprecedented growth and are expected to approach 60TB native (more than 120TB compressed) by 2019.
  • Tape WORM (Write Once, Read Many) and the growing requirement for encryption address media data protection for the compliance and archival markets.
  • The capital expense (capex) per gigabyte for automated tape library is expected to remain below that of magnetic disk for at least the next 10 years.
  • The operating expense (opex) for automated tape libraries should remain significantly below disk storage as people, facilities, and energy costs rise. The average tape library administrator manages from 40TB to more than 1PB (varies based on the capacity of a single library) of data. Storage administrators currently manage an average of nearly 30TB for non-mainframe disk storage and more than 100TB for mainframe disk storage.
  • The promise that network bandwidth would be sufficient to replace physical transport for moving large amounts of data has yet to materialize. The growth of backup data continues to exceed the rate of growth of affordable bandwidth. This trend is expected to continue for the foreseeable future.
  • Tape drives and media have a much longer storage life than disk drives, eliminating the need for technology conversions as frequently as disk.

Potential enhancements may include new innovations such as:

  • Uniquely addressable tape partitions, high-performance regions, data tags, metadata, and indexing schemes such as Extensible Access Method (XAM) for faster retrieval
  • A tighter integration with magnetic disk arrays, cartridge-based flash memory storage, and flash/RFID chips to offset existing tape limitations in access times (time to first byte of data), signaling a new and potentially revolutionary approach to accessing data on tape
  • A tighter integration with databases and other applications for enhanced archival functionality.

Despite 15 years of characterization as a dinosaur, tape isn’t likely to go away any time soon. In difficult economic times, the value of using a tiered storage strategy with more economical tape for optimal tier 3 solutions couldn’t be greater. Remote storage, data vaults, and the ability to move data when electricity isn’t available favor tape, as optical disc is no longer a data center technology creating additional opportunity. Tape has a promising technological future, but the question remains as to whom will become the primary driver of tape? Z