Computational Storage Drive Patents: Leaders & White Space 2026
Filing growth compares 2021 (23 records) with 2024 (29) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 261 records in scope (CR5), not by the ranked leaders only.
What the computational storage drive patent record shows
A computational storage drive moves processing closer to the data it stores, embedding compute logic inside the drive controller rather than shipping raw data to a host CPU. The 261 records in this dataset span filings from 2015 through the current cut-off, covering architectures, offload accelerators, and the software interfaces that let a host request in-drive processing. The field sits at the intersection of storage engineering and general computing, which shows up directly in the IPC mix: G06F electric digital data processing covers 69.0% of records, but no single class dominates the remainder.
Filing activity is concentrated: a single assignee holds 41 records against a ranked field of 100 companies, and the top five together account for 32.6% of all 261 records in scope. That leaves a long tail of single- or few-filing entrants below the leaders, which is where freedom-to-operate questions tend to cluster.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
Filing trend and technology composition
The two views below track how filing volume has moved since the field's early records and how patent activity distributes across the IPC subclasses these documents touch.
Filing trend: growth through the last complete year
Annual filings rose from 2 records in 2017 to a peak of 29 in 2024, with the 2021-to-2024 span showing +26% growth (23 to 29). 2025 and 2026 figures in the raw trend appear lower only because publication lags filing by roughly 18 months; they should not be read as a slowdown.
Technology composition across IPC subclasses
G06F (69.0% of records) confirms this is fundamentally a digital-data-processing field, but meaningful shares sit in G06N AI-model computing (8.0%), A61B diagnosis and surgery (7.7%), H04L digital transmission (5.7%), G11C static memories (5.4%), G11B magnetic/optical storage (5.0%), G06Q business data processing (4.2%) and G06T image processing (3.8%). Because records can carry multiple classes, these shares sum to more than 100%, and they show breadth of application rather than a single dominant sub-technology.
Shares are the percentage of the 261 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on High-Performance Storage: Computational Storage Drive Patent Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about high-performance storage: computational storage drive patent landscape and every answer comes back with the patent numbers behind it.
Try EurekaA representative filing and the most-cited prior art
Multimedia on demand method, apparatus, system, computational storage drive and storage medium
A multimedia on demand system includes a host and a computational storage drive. The host sends a request for multimedia on demand to the drive; the drive performs an on-demand processing process including flow control and either a codec process or a multiplexing process, then returns the result to the host.Filed by Samsung Electronics, this 2026 application shows the field's current direction: pushing codec and multiplexing workloads into the drive itself rather than treating computational storage as a pass-through cache layer.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20060253894A1 | Mobility device platform | 341 |
| 2 | US6567889B1 | Apparatus and method to provide virtual solid state disk in cache memory in a storage controller | 252 |
| 3 | US5257367A | Data storage system with asynchronous host operating system communication link | 236 |
| 4 | US6507672B1 | Video encoder for digital video displays | 196 |
| 5 | US20200150508A1 | Building network | 96 |
| 6 | US20170153290A1 | Systems and related methods for determining self-discharge currents and internal shorts in energy storage ce… | 91 |
| 7 | US20030139907A1 | System, Method, and Product for Nanoscale Modeling, Analysis, Simulation, and Synthesis (NMASS) | 78 |
| 8 | US20120143285A1 | Handheld excitation terminal and EMF emitter providing dynamic optimization of emission and therapeutic effec… | 76 |
| 9 | US6839750B1 | Single management point for a storage system or storage area network | 69 |
| 10 | US9354927B2 | Securing virtual machine data | 45 |
Citation counts favour older records simply because they have had longer to accumulate citations inside the searched corpus; treat them as a signal of influence on the field, not as a ranking of current importance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the numbers mean for filing strategy
Three patterns in the dataset matter more than the raw counts: how concentrated ownership already is, where growth is real versus an artifact of publication lag, and which IPC branches carry filing density versus which stay thin.
Ownership is concentrated at the very top
The leading assignee alone holds 41 records, and the next four combined add only 44 more. Below the top five, the ranked field of 100 companies thins quickly into single- and few-filing entrants, which is where most freedom-to-operate risk in this field actually sits.
Real growth, not a publication artifact
Filings rose from 23 records in 2021 to 29 in 2024, the last year the dataset treats as complete. Counts for 2025 and 2026 look lower only because publication lags filing by roughly 18 months; they will fill in as more applications publish.
Core storage architecture still anchors the field
G06F electric digital data processing appears in more than two-thirds of records, but secondary classes in AI-model computing, digital transmission and even diagnosis-and-surgery equipment show computational storage claims reaching into adjacent application domains.
Filing here is overwhelmingly solo
Only two co-assignee pairs appear across the dataset, both involving the same organisation. Most applicants in this field file independently rather than through joint ventures or research consortia.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to high-performance storage: computational storage drive patent landscape, with the prior art for and against each one.
Where to take this analysis
The dataset points to a concentrated top and a long tail below it. The next step is deciding whether to file around the leader's claim space or move toward the thinner IPC branches where coverage is still open.
Map freedom-to-operate against the top holders
With one assignee at 41 records and the top five at 32.6% of the field, a targeted claim-chart review of their most recent families is a faster way to spot conflict than scanning the full 261-record set.
Explore assignee claims in EurekaWatch the under-claimed IPC branches
G06Q, G06T and G11B each sit under 6% of records. If your architecture touches business-data offload or image processing inside a drive controller, that thinner coverage is worth checking before assuming the space is occupied.
Run a white-space search in EurekaTrack filings past the 2024 peak
Because 2025-2026 counts are still filling in, re-running this trend in a few months will show whether growth past 29 records a year continues.
Set up a filing alert in EurekaCommon questions on computational storage drive patents
One assignee holds 41 of the 261 records in this dataset, making it the clear leader by a wide margin over the rest of the field. The next four leaders together add 44 more records, bringing the top five to 32.6% of all filings in scope. Below that, the ranked list of 100 companies thins out quickly into a long tail of entrants with only a handful of filings each, so most competitive risk sits outside the top names rather than concentrated within them.
Filings grew from 23 records in 2021 to 29 in 2024, a 26% increase, and 2024 is the most recent year the dataset treats as complete. Counts appear to dip in 2025 and 2026, but that is a publication-lag artifact: filed applications typically take roughly 18 months to publish, so the newest years are always undercounted at the time of any search. Read the trend through 2024, not through the tail years.
G06F, the IPC subclass for electric digital data processing, appears in 69.0% of the 261 records, confirming that core storage-and-compute architecture is still the field's centre of gravity. Secondary classes are much smaller: G06N for AI-model computing sits at 8.0%, A61B for diagnosis and surgery equipment at 7.7%, and H04L for digital transmission at 5.7%. Because a single record can carry several IPC classes, these percentages add up to more than 100%, and they reflect breadth of application rather than competing dominant technologies.
The thinner IPC branches in this dataset are G06Q business-data processing (4.2% of records), G06T image processing (3.8%), and G11B magnetic/optical storage (5.0%), all well below the 69.0% carried by core G06F claims. These lower shares do not prove the technology is immature, only that fewer claims currently occupy that space, which is exactly where a new filing has more room to stand on its own merits. A practitioner targeting these branches should still run a full clearance search rather than relying on subclass share alone.
US20260052282A1, filed by Samsung Electronics and published 2026-02-19, describes a computational storage drive that performs codec and multiplexing processing for multimedia-on-demand requests directly on the drive, rather than passing raw data back to the host. It illustrates the current direction in the field: moving specific, well-defined processing tasks into the drive controller itself. Reviewing filings like this one against your own architecture is a practical way to check for claim overlap before committing engineering resources.
Research High-Performance Storage: Computational Storage Drive Patent Landscape in depth with Eureka
Go past this page: query the whole high-performance storage: computational storage drive patent landscape corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.