Tape Remains Important Component of Data Management Strategies

By Daniel Jan

Have you heard of the “ zettabyte apocalypse ?” It’s a term coined to describe how organisations must be prepared to increase storage to an additional 40 to 60 zettabytes of data (per most analysts’ predictions) over the next four years or risk being left susceptible to the type of “apocalyptic” disasters—compliance violations and fines, data loss and theft, etc.—that could run them out of business.

Fortunately, however, tape storage—and most notably, its higher-than-ever level of capacity—is a key component in helping organisations prepare for the coming data influx. According to an April 2017 report from the LTO Program Technology Provider Companies, “a record 96,000 petabytes1 (PB) of total compressed tape capacity shipped in 2016, an increase of 26.1 percent over the previous year.” LT-07, one of the latest iterations of tape, now leads the industry in reliability; its capacity and data rates are growing faster than other storage options; and it is one of the most energy efficient options on the market.

Here’s a few more reasons why more data than ever is being stored on tape:

Reliability - Bit Error Rate (BER) is a way to express the effectiveness of reading data from a disk or tape drive. It’s a measure of the total number of expected erroneous bits as compared to the total number of bits received, and it serves to quantify the likelihood of a faulty bit. LTO-7 is rated at one bit in error per 1×1019 bits read, scoring 1,000 times higher than top-rated HDDs, which are currently rated at one bit in error per 1×1016 .   This means the likelihood of even the top-rated disk writing an incorrect bit is 1000x greater than tape.

Capacity - Tape continues to grow its capacity at a greater rate than other storage technologies. LTO-8 plans to deliver 12.8TB native capacity and the next iterations are aiming for native capacities of more than 15TB. Tape also benefit from a 2.5x compression boost, as most of the data written to tape is compressed. LTO-7 transfers data at 300MB/sec and the TS1150 at 360MB/sec, which is significantly faster than the typical 7,200RPM HDD at 160MB/sec.

Energy Consumption - One of the greatest benefits of tape is that when it’s not in use (i.e., when data is not being transferred to it) it sits virtually dormant. Not only does that result in greater reliability than disk, which is powered on and is therefore more susceptible to downtime, but also leads to significantly less energy consumption. Most industry estimates place the total energy consumption of tape at 5% that of disk. Today, as most organisations strive to become “greener” and more energy efficient, tape is proving to be the most logical storage choice.

Media Life - Tape’s longevity is simply unbeatable. LTO tape has a life span of 30 years. By comparison, disk is typically operational from 3 to 5 years before replacement, and LTO tape drives can read the current version and the two prior LTO versions thereby minimising the conversion effort.

Tape is Built for the Future - While most analysts predict an additional 40 to 60 zettabytes to the digital information burgeon over the coming four years, what about the years after that? Will the amount of data continue to rapidly increase?  Or will growth eventually slow? Well, if a new report from IDC is any indication, a growth of 40 to 60 zettabytes will be little more than drop in the proverbial bucket. Recently, they predicted the total amount of digital data created worldwide will approach 180 zettabytes by 2025.

So, how is an organisation to prepare? Disk storage is no longer growing its capacity at a sufficient rate or with the affordability characteristics that will make it suitable for zettabyte storage, and, as of right now, that much of a capacity seems a stretch even for tape. Fortunately, though, planned innovations in tape storage can help organisations future-proof against the coming data boon.

Here’s how:

Lowering cost - Tape is the most cost-effective technology for long-term data retention. Capacity can scale without adding more drives, unlike HDDs, and some industry estimates place the Total Cost of Ownership (TCO) for tape about six times lower than equivalent HDDs systems. To walk through the various ways tape proves to be one of the more cost-effective storage options on the market is a task far beyond the scope of this article. Fortunately, this report by the Clipper Group offers an exhaustive breakdown of tape’s costs as compared to disk and cloud solutions.  It’s worth a read, but if you want to skip to the end, they conclude, “Tape library solutions still have a significant economic advantage over disk-based solutions on a cost per terabyte stored basis.” Really, the bottom line is: As the amount of data an organisation must store grows, tape gets cheaper and cheaper.

Using LTFS software - Linear Tape File System (LTFS) enables direct, intuitive and graphical access to data stored on LTO tapes, thereby eliminating the need for additional tape management and proprietary software to access data.

Introducing Active Archives - A combined solution of open systems software, disk and tape hardware, active archives give users an automated way to store and manage all data across multiple storage types (HDDs, tape, and cloud storage). It improves tape access time by serving as a cache buffer for a tape library, and enables a high percentage of accesses to the tape subsystem to be satisfied from HDDs, thereby avoiding physical tape access and making it well suited for higher-performance and large capacity and archive applications.

The emergence of tape as NAS - Tape as NAS integrates an LTO tape library with a front-end NAS and LTFS to deliver a higher performance, scalable archive solution. A tape library as NAS enables users to leverage familiar file system tools, and even drag and drop files directly to and from a tape cartridge, just like a disk-based NAS.

Tiered Storage - Tiered storage uses standard HSM software functionality that enables the storage administrator to define policies for data migration and retention to control the movement of petabytes of data from more expensive HDD storage devices to less-expensive tape storage.

HPC Embracing HPSS and Tape - HPSS (High Performance Storage System) software is widely used in the HPC market providing highly flexible and scalable hierarchical storage management functionality that optimises large-scale storage resources by keeping recently used data on disk and less recently used data on cost-effective tape. In addition, HPSS enables RAIT (Redundant Array of Inexpensive Tape) effectively multiplying the data rate and improving the availability of tape subsystems.

Tape for cloud - One large misconception is that by choosing tape, you must forgo cloud. But, in reality, they can be complementary.  Using tape for cloud archives, rather than HDDs, greatly reduces cloud TCO. The role of tape in the cloud will continue to grow, as cloud providers seek to lower their storage costs and relieve pressure from exclusively using more costly HDDs for lower activity and archival data. For more, “Why Tape Rules: Using Tape Backup in the Cloud” provides an excellent overview on how these technologies can work together.

Takeaways - Despite the prevailing media death knells, steady advancements have made tape the go-to backup and archiving choice for organizations worldwide. It has surpassed disk in key categories like capacity, reliability, and total cost of ownership, and has proven to work alongside cloud effectively. Most importantly, though, it’s what’s needed to survive the “zettabyte apocalypse.”

Dan Jan is Principal for Tape Product Management at Iron Mountain, Inc. With over 25 years of industry experience spanning semiconductor memory to hard disks, optical, and tape storage technologies, Jan also held leadership roles at startups in Silicon Valley, where he spent 12 years.