LTO Tape Roadmap Targets 365TB Cartridges as Archive Demand Grows

LTO tape is still gaining capacity, but its roadmap now puts more weight on operating trade-offs. The current LTO-10 format holds up to 40TB of uncompressed data per cartridge. HPE, IBM and Quantum, which run the LTO Program, now project 365TB for LTO-14, below the 576TB target they outlined in 2022. Their revised plan reflects a practical constraint: storing more data on each cartridge can also mean longer waits to write or recover it.

More capacity through denser and longer tape

LTO has come a long way since its 2000 launch. LTO-1 stored 100GB and transferred data at 20 MB/s. Capacity rose through successive generations: 200GB for LTO-2, 400GB for LTO-3, 800GB for LTO-4, 1.5TB for LTO-5, 2.5TB for LTO-6, 6TB for LTO-7, 12TB for LTO-8 and 18TB for LTO-9.

The first LTO-10 cartridges, introduced in 2025, hold 30TB natively. An extended-length version later raised that figure to 40TB. Across roughly 25 years, LTO’s capacity per cartridge has grown 400-fold without a change to the format’s form factor. By comparison, 3.5-inch hard drives grew from about 75GB to as much as 44TB over a similar period, a roughly 585-fold increase. Tape has scaled more slowly, while avoiding major increases in internal complexity.

LTO-10’s capacity comes from two changes: recording more data in a given area and fitting more tape inside the cartridge. The format records 15,104 data tracks, compared with 8,960 for LTO-9, and uses higher linear density. A barium ferrite recording layer supports the denser recording. The head also relies on a redesigned, high-bandwidth servo system and a low-mass mechanism to track narrower paths accurately, including when equipment is exposed to shock or vibration.

Both LTO-10 cartridge sizes use a strontium-doped barium ferrite layer. The 40TB version adds a thinner, smoother Aramid base film, allowing more tape to fit in the same cartridge. MagStor puts the tape length at about 1,335 meters, up from approximately 1,035 meters in the 30TB cartridge. The 40TB media works in existing LTO-10 drives.

Laboratory work points to headroom beyond current products. In 2020, IBM and Fujifilm demonstrated strontium ferrite tape with an areal density of 317 Gb/in² and said it could theoretically support a 580TB cartridge. The prototype used 56.2nm-wide tracks and a 29nm TMR read sensor. That result signals technical potential; it does not establish a commercial LTO capacity or release schedule.

A slower roadmap, with recovery time in view

The LTO Program’s 2022 roadmap projected roughly doubled capacity with each generation: 72TB for LTO-11, 144TB for LTO-12, 288TB for LTO-13 and 576TB native for LTO-14, or 1.44PB with compression. In November 2025, the companies replaced those figures with targets of 70TB, 120TB, 210TB and 365TB, respectively.

The revised targets don't prove that tape has reached a physical limit. The consortium says it is balancing capacity against reliability, affordability, efficiency and cost per terabyte. It also argues that cartridges with less maximum capacity can take less time to fill and recover. That trade-off matters for archival systems, where a high capacity figure is useful only if the time required to retrieve data fits the operator’s needs.

Bandwidth remains a constraint. LTO-8 supports 360 MB/s natively, LTO-9 reaches 400 MB/s, and LTO-10 stays at 400 MB/s. Compression can raise effective throughput to as much as 1,200 MB/s when data compresses at the assumed 2.5:1 ratio. From LTO-8 to LTO-10, maximum native capacity rose from 12TB to 40TB, while native transfer speed improved by about 11%. At 400 MB/s, writing 40TB in sequence would take roughly 28 hours under ideal conditions, before overhead. The consortium has not published transfer-rate targets for later generations.

Migration is the cost of LTO-10’s redesign

LTO-10 also ends the format’s backwards compatibility. Through LTO-7, a new drive could generally read cartridges from two generations earlier and write one generation back. LTO-8 and LTO-9 narrowed that support to the previous generation. LTO-10’s redesigned head removes it: an LTO-10 drive reads and writes LTO-10 media only. Its 30TB and 40TB cartridges are interchangeable with each other.

For organizations maintaining archives over decades, adopting LTO-10 means either migrating data from LTO-8 or LTO-9 media, or keeping older drives available. That choice affects the cost and logistics of a transition, especially when archives must remain retrievable for years.

Why tape still fits archive systems

Shipments reached 160.3 exabytes in 2025, down 9% year over year. The LTO consortium reported a 57% year-over-year rise in capacity shipments in the first quarter of 2026, attributing the increase to AI data growth, archival demand, cyber-resilient backup needs, continued LTO-9 adoption and the start of the LTO-10 transition.

AI systems generate datasets, checkpoints, models, logs and other files that may need to be retained without staying on higher-cost SSDs or hard drives. Tape offers high capacity at a low cost per terabyte, uses almost no power while cartridges are offline, and supports hardware encryption. Removing a cartridge from its drive creates a physical air gap from network-based attacks. LTO’s LTFS file system also stores indexing information on tape, helping users manage archived files.

The trade-off is access speed: tape is sequential, so it can't provide the random access of SSDs or hard drives. For data that can be retrieved in sequence and does not need constant access, that limitation may be acceptable. LTO’s path from 100GB cartridges to 40TB media, along with its 365TB LTO-14 target, suggests the format will continue to serve that role. The next test is whether each capacity increase can fit operators’ needs for reliability, cost and recovery time.