Realizing Your Long-Term Data Strategy
Organizations are retaining more data for longer periods, often because of regulatory requirements, research, historical records, security needs and the growing value of data for analytics and AI. As those archives grow into the tens or hundreds of petabytes, storage capacity is only part of the challenge.
The bigger challenge is identifying whether that data can still be accessed and moved efficiently when the underlying technology changes.
That issue is becoming especially relevant with LTO-10. According to the LTO Program, LTO-9 drives can read LTO-8 and LTO-9 media, while LTO-10 drives can read and write only LTO-10 media.
For organizations with large LTO-8 archives, that means moving to LTO-10 is not simply a matter of installing new drives. Existing data must first be read through compatible infrastructure and migrated into the new environment.
For a small archive, that may be manageable. At petabyte scale, it can become a significant infrastructure project.
The Migration Math Changes the Conversation
Consider an organization with 100 PB of retained LTO-8 data. At a native transfer rate of approximately 360 MB/s per source drive, 10 drives operating continuously would provide about 3.6 GB/s of aggregate throughput.
At that rate, reading 100 PB would take roughly 321 days… and that is a theoretical calculation.
Real migrations must account for media handling, verification, retries, software overhead, competing workloads and the performance limitations of the rest of the environment. The source drives are only one part of the path. Libraries, servers, storage fabrics, networks, migration software and target systems all have to support the required throughput.
The U.S. Geological Survey highlights this issue in its FY 2025 Offline Archive Media Trade Study, noting that transfer performance directly affects how quickly migration and verification can be completed.
That creates an important distinction for long-term storage planning: Having enough capacity to retain the data does not necessarily mean you have enough capacity to migrate it.
Storage Media Can Outlive the Technology Around It
Long-term archives may need to retain information for decades, while the systems used to access that information evolve much more quickly. The Library of Congress actively monitors its storage systems and media for obsolescence and plans migrations to new media and platforms as part of long-term digital preservation.
That is an important lesson for commercial and government organizations alike.
The physical life of the media is not the same as the operational life of the infrastructure required to read it. Compatible drives, interfaces, software and technical expertise can all become harder and more expensive to maintain over time. That is why migration planning should happen while the current environment is still fully supportable, not when aging infrastructure creates an urgent deadline.
The Answer Is Not Automatically Tape, Flash or Cloud
LTO-10’s compatibility change does not mean tape no longer has a role in long-term storage. For many workloads, tape continues to provide strong economics, storage density and offline protection. In fact, after evaluating several technologies, USGS continued to recommend LTO for its own offline archive requirements.
Other data may be better suited to flash, disk, object storage or cloud platforms because of performance, accessibility, recovery or analytics requirements.
For many organizations, the right answer will be a combination of technologies. The more useful question is not simply, “What should replace our current storage?”
It is, “Which data belongs on which storage tier, and how difficult will it be to move that data when requirements or technologies change?”
That assessment should consider migration time, access requirements, recovery objectives, data growth, hardware availability, security, operational resources and long-term economics.
Long-Term Retention Creates a Migration Obligation
Organizations often evaluate storage by asking how much data they have, how quickly it is growing and how much it costs to retain. Those questions remain important, but long-term retention creates another obligation.
The organization must be able to keep that data accessible, protected and movable throughout its lifecycle.
For large archives, that means understanding which generations of media exist in the environment, how much data resides on each, what hardware is required to access it and how long a full migration would realistically take.
It also means accounting for continued data growth while that migration is happening.
If an archive takes a year to migrate while significant new data continues to accumulate, the migration window itself becomes a strategic consideration.
Before Your Next Refresh, Run the Math
Storage modernization should begin with the data, not the product. Before making the next archive or storage decision, organizations should understand what they are retaining, how long it must remain available, how quickly it needs to be accessed and what would be required to move it.
For organizations managing significant amounts of retained data, one question can quickly expose whether there is a larger issue: If you had to migrate your data today, how long would it actually take?
If the answer is measured in months or years, migration should already be part of the data lifecycle strategy.
For more than 35 years, Jeskell has helped Federal and commercial organizations design and modernize storage environments around the requirements of their data. By evaluating existing infrastructure, data growth, retention needs, accessibility requirements and migration constraints, organizations can make better decisions about the mix of technologies that will support their data over the long term.
Before your next storage refresh, run the migration math. Understanding how long it would take to move your data today could change where you decide to store it tomorrow.