Fastest
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 →https://www.patsnap.com/resources/blog/rd-blog/remote-direct-memory-access-networking-patent-landscape/ · Patsnap · data cut-off 2026-07-31 · downloaded from the live page
Patent Landscape · High-Performance InterconnectRemote Direct Memory Access Networking Patents
- 42.8% concentration. The five leading filers together hold 1,936 of 4,527 records in scope — nearly half the field sits with a handful of companies.
- Filing cooled after a 2021 peak. Filings hit 446 in 2021 and stood at 314 by 2024, a 30% pullback over that span, even before accounting for publication lag.
- Two IPC classes carry almost everything. G06F and H04L alone cover 74.5% and 48.1% of records respectively; the remaining six classes each sit under 2%.
4,527
Published Records
43%
Top-5 Share of All Records
-30%
Filing Growth 2021→2024
US
Leading Jurisdiction
Filing growth compares 2021 (446 records) with 2024 (314) — 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 4,527 records in scope (CR5), not by the ranked leaders only.
Published byPatsnap Research··6 min readSourced from Patsnap Eureka
Field Overview
What this landscape covers
This dataset tracks patent activity built around remote direct memory access (RDMA) transport mechanics: queue pair management, memory registration, completion queue handling, kernel-bypass data paths, lossless fabric behaviour, retransmission handling and connection state tracking. It spans 4,527 published records filed or published between 2015 and mid-2026, drawn from a search string that anchors on RDMA-specific transport terms rather than general networking language.
The scope favours transport-layer and adapter-level claims over application-layer uses of RDMA, which is why digital data processing and digital transmission classes dominate the composition figures while wireless, video and memory-hardware classes appear only at the margins.
Filing activity and technology composition, 2015–2026
Source: Patsnap Eureka. Assignee ranking and totals. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
10,000 free credits to start
The Numbers
Filing trend and technology composition
Two views of the same 4,527-record dataset: how filing volume has moved year over year, and which IPC subclasses the claims actually sit in.
Filings rose to a 2021 peak, then eased
Annual filings climbed from 211 in 2017 to a peak of 446 in 2021, then declined to 314 by 2024 — a 30% drop over that three-year window. 2025 and 2026 figures are still incomplete because publication typically lags filing by around 18 months, so the most recent years will fill in as more records publish.
Two classes dominate the claim space
G06F (electric digital data processing) appears in 74.5% of records and H04L (digital information transmission) in 48.1%, reflecting RDMA's position at the intersection of data-path processing and network transport. Multiplex communication, static memory, wireless, imaging and video classes each account for under 2% of records, since a record can carry multiple IPC codes these shares add to more than 100%.
Shares are the percentage of the 4,527 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Source: Patsnap Eureka. Filing trend and technology composition. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.
Go deeper on Remote Direct Memory Access Networking with Eureka
This page is one run against one query. Ask Eureka your own question about remote direct memory access networking and every answer comes back with the patent numbers behind it.
Try EurekaPrior Art Signals
Most-cited records and a representative filing
Representative Filing
US20160212214A1 — Tunneled remote direct memory access (RDMA) communication
AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD.
Tunneling packets of one or more RDMA unreliable queue pairs of a first adapter device through an RDMA reliable connection by using reliable and unreliable queue context stored in the first adapter device. The reliable connection is initiated between two RDMA reliable-connection queue pairs, one on each adapter, with context maintained per queue pair type.Filed by Avago Technologies (Singapore), 2016-07-21 — illustrates how tunneling unreliable queue-pair traffic through a reliable connection was claimed at the adapter level.
Highest-cited records in scope
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20060136570A1 | Runtime adaptable search processor | 704 |
| 2 | US7010607B1 | Method for training a communication link between ports to correct for errors | 684 |
| 3 | US20040165588A1 | Distributed network security system and a hardware processor therefor | 651 |
| 4 | US20050108518A1 | Runtime adaptable security processor | 528 |
| 5 | US7685254B2 | Runtime adaptable search processor | 468 |
| 6 | US20050018669A1 | Infiniband subnet management queue pair emulation for multiple logical ports on a single physical port | 437 |
| 7 | US20130117766A1 | Fabric-Backplane Enterprise Servers with Pluggable I/O Sub-System | 403 |
| 8 | US20130151646A1 | Storage traffic communication via a switch fabric in accordance with a VLAN | 392 |
| 9 | US20040010612A1 | High performance IP processor using RDMA | 391 |
| 10 | US8627136B2 | Non-disruptive failover of RDMA connection | 388 |
Citation counts accumulate over time, so older records lead this table; treat them as markers of influence on later filings rather than as current state of the art.
Each row carries its publication number; clicking a row searches Eureka by that number.
Source: Patsnap Eureka. Citation counts and representative records. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Run it yourself
Put your own technology through the same analysis
Where to run it
For builders
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 →Analysis
What the concentration and momentum figures mean
Filing share and recent-year activity point to a field where the largest holders staked claims early and have since slowed, while the underlying transport mechanics remain widely referenced.
Concentration
42.8%
of 4,527 records held by top 5
Half the field sits with a handful of filers
The five leading assignees combined hold 1,936 of the 4,527 records in scope, and the ranked ten extend that to 57.4%. For a new entrant, this means core queue-pair and memory-registration claims are likely already occupied by one of a small number of holders.
Based on the 100-company ranked assignee list.
Momentum
-30%
filings, 2021 to 2024
Peak filing has passed, not activity itself
Filings peaked at 446 in 2021 and had fallen to 314 by 2024. Several of the largest historical filers show sharp year-on-year declines in the latest year, but partial and lagging publication data make it too early to call this a retreat from the technology.
2025–2026 figures are still filling in due to publication lag.
Composition
74.5%
of records touch G06F
Claims cluster on data-path processing
Nearly three-quarters of records carry a G06F classification and just under half carry H04L, confirming that most patent activity addresses how data moves through the adapter and transport stack rather than at the application or wireless edge.
Shares exceed 100% combined because records can carry multiple IPC codes.
Eureka AI Agent
Looking for what nobody has claimed yet?
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to remote direct memory access networking, with the prior art for and against each one.
Source: Patsnap Eureka. Co-assignee relationships and derived observations. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Competitive Landscape
Who holds the field, and where it is thinning out
The ranked leaders include established server, silicon and hyperscaler names, but recent-year filing counts have dropped sharply across nearly all of them, suggesting the intense claim-building phase for core RDMA transport has passed even as adjacent branches remain open.
Leader
522
records
A single filer holds the largest share
The top-ranked assignee's 522 records outpace the fifth-place holder's 211 by more than double, indicating a genuine leader rather than a tight cluster at the very top.
Counted in patent families / records.
Mid-field
83
records at 10th place
A steep drop after the top five
Filing counts fall from 211 at fifth place to 83 at tenth, showing the concentration is front-loaded rather than spread evenly across the ranked group.
From the 100-company ranked assignee list.
Recent activity
-92% YoY
at two leading filers
Even active filers have pulled back sharply
Several of the largest historical filers, including leading networking and semiconductor names, show year-on-year declines of 80% or more in the latest year, alongside others reporting zero filings in that period.
Latest-year figures are partial due to publication lag.
Filing momentum by assignee, latest year
| Assignee | Recent year | YoY |
|---|---|---|
| Huawei Technologies Co., Ltd. | 2 | -92% |
| Mellanox Technologies Ltd. | 1 | -92% |
| Microsoft Technology Licensing, LLC | 1 | -80% |
| International Business Machines Corporation (IBM) | 0 | -100% |
| Intel Corporation | 0 | -100% |
| Samsung Electronics Co., Ltd. (Korea) | 0 | -100% |
| Oracle International Corporation | 0 | -100% |
| Western Digital Technologies, Inc. | 0 | — |
Source: Patsnap Eureka. Assignee-level momentum. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Next Steps
Where to take this analysis
The dataset points to a field with a settled leadership tier and a few technical corners still open. The next steps depend on whether the goal is freedom-to-operate, licensing strategy, or identifying a filing gap.
Check freedom-to-operate against the leaders
With 42.8% of records held by five assignees, any new queue-pair or memory-registration claim should be checked against that group's portfolios first.
Explore assignee portfolios in EurekaTrack the post-2021 filing pattern
The shift from 446 filings in 2021 to 314 in 2024 is worth monitoring as later years publish, to see whether it reflects consolidation or a genuine slowdown in transport-layer claims.
Set up filing trend alerts in EurekaScope the under-claimed branches
Wireless, imaging and memory-hardware classes each sit under 2% of records, which may reflect either low technical relevance or genuine white space worth a closer claim search.
Run a white-space search in EurekaSource: Patsnap Eureka. Forward-looking reading of the same dataset. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
FAQ
Common questions about RDMA networking patents
The field is led by a single assignee holding 522 of the 4,527 records in scope, well ahead of the fifth-ranked holder at 211. The top five assignees combined account for 42.8% of all records, and the top ten reach 57.4%. This concentration includes established server, networking-silicon and hyperscaler companies, meaning most core queue-pair, memory-registration and completion-queue mechanics are already covered by a small group of portfolios.
Filings rose from 211 in 2017 to a peak of 446 in 2021, then declined to 314 by 2024 — a 30% drop over that three-year span. Because publication typically lags filing by around 18 months, the 2025 and 2026 figures in this dataset are still incomplete and should not be read as evidence of a continued decline. The honest read is that the intense early-2020s filing surge has passed its peak, not that interest in the technology has ended.
Three-quarters of records (74.5%) carry a G06F classification for electric digital data processing, and just under half (48.1%) carry H04L for digital information transmission, reflecting RDMA's core position between data-path processing and network transport. Smaller classes covering multiplexing, static memory, wireless networks, imaging and video each account for under 2% of records, indicating these are peripheral rather than central to how RDMA has been claimed.
The classes furthest from the dominant G06F/H04L cluster — wireless communication networks, video/pictorial communication, and static memory hardware — each cover well under 2% of the 4,527 records, which points to comparatively open claim space around RDMA-over-wireless paths, accelerator-attached memory registration, and connection-state handling during adapter migration. These branches are under-claimed relative to core transport mechanics, though a proper freedom-to-operate check is still needed before filing. Low density in a class does not by itself confirm the technology is either commercially viable or unclaimed elsewhere outside this search scope.
US20160212214A1, filed by Avago Technologies (Singapore) in 2016, claims a specific method of tunneling unreliable RDMA queue-pair traffic through a reliable RDMA connection using separate reliable and unreliable queue contexts stored on the adapter. Anyone implementing a similar tunneling approach — routing unreliable-datagram-style traffic over a reliable-connection transport at the adapter level — should review this filing's specific claim language around context storage and queue-pair initiation. It does not block RDMA implementations generally, only this particular tunneling architecture and its close variants.
Answers are grounded in the same dataset. Derived from a Patsnap search on Remote Direct Memory Access Networking covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Research Remote Direct Memory Access Networking in depth with Eureka
Go past this page: query the whole remote direct memory access networking 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.