Container Orchestration Patents: Who Leads, Trends 2026
- Filing peaked in 2023 at 37 families, then flattened — the midpoint year (2022, 16 families) sits well below peak, so growth has already crested rather than accelerated.
- G06F carries 93 of 103 records and H04L 43, meaning almost every filing sits at the intersection of core compute scheduling and network transmission claims, not in a separate niche.
- The United States receives 68 of 103 filings, more than four times the next office (EPO, 15), so US prosecution and prior art dominate freedom-to-operate analysis here.
Filing growth compares 2021 (15 records) with 2024 (9) — 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 103 records in scope (CR5), not by the ranked leaders only.
What this patent set covers
This landscape tracks patent families whose titles reference container orchestration, cluster scheduling or Kubernetes scheduling, filtered to claims and descriptions naming bin packing, resource isolation, autoscaling, multi-tenancy or cold start behaviour. The IPC filter (G06F9, H04L67, G06F11) keeps the set anchored in program execution, network-application-layer transmission and error-recovery/testing classifications rather than general cloud infrastructure.
103 patent families were published between 2015 and the mid-2026 data cut-off. Because publication typically lags filing by around eighteen months, the 2025-2026 counts in any trend chart understate actual filing activity for those years.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
Filing trends and technology composition
Two views of the same 103-family dataset: how filing volume moved year over year, and which IPC subclasses carry the claim weight.
A single peak year, not a sustained climb
Filings sat at zero in 2017, built steadily, and reached a peak of 37 families in 2023 before falling back. With 2022 at only 16 — the rough midpoint of the series — the growth curve front-loaded its rise into 2020-2023 and has since flattened or declined, consistent with a technology whose foundational claim space is now largely staked out.
Concentrated in compute and transmission classes
G06F (electric digital data processing) appears in 93 of 103 records and H04L (digital information transmission) in 43, confirming that orchestration and scheduling claims are drafted primarily as compute-execution patents with network-transmission elements layered in. AI-related classifications (G06N, 6 records) and wireless (H04W, 4) are present but marginal, suggesting AI-driven scheduling is still an early rather than dominant claim strategy.
Shares are the percentage of the 103 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Container Orchestration and Scheduling with Eureka
This page is one run against one query. Ask Eureka your own question about container orchestration and scheduling and every answer comes back with the patent numbers behind it.
Try EurekaThe most-cited records and a representative filing
Multi-tenancy interference model for scaling in container orchestration systems (US20230104787A1, SAP SE, 2023-04-06)
Methods, systems, and computer-readable storage media for an autoscaling framework that uses a multi-tenant interference model for selectively scaling resources in container orchestration systems based on an estimated performance, an actual performance, and a violation count of each resource unit of a plurality of resource units. In some implementations, the autoscaling framework of the present disclosure mitigates length of response times in applications composed of a set of services.Filed by SAP, this record ties autoscaling decisions to a per-tenant interference model rather than raw utilisation thresholds — a claim structure that narrows what a competing scheduler can call 'autoscaling' without infringing.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20180300174A1 | Efficient queue management for cluster scheduling | 97 |
| 2 | US20210311764A1 | Container orchestration in a clustered and virtualized computer system | 89 |
| 3 | US20210034423A1 | Container orchestration in decentralized network computing environments | 49 |
| 4 | US20230104368A1 | Role-based access control autogeneration in a cloud native software-defined network architecture | 40 |
| 5 | US20200351650A1 | Blockchain-based front-end orchestrator for user plane network functions of a 5g network | 34 |
| 6 | US20220329651A1 | Apparatus for container orchestration in geographically distributed multi-cloud environment and method using … | 32 |
| 7 | US20210397465A1 | Container-as-a-service (CAAS) controller for monitoring clusters and implemeting autoscaling policies | 31 |
| 8 | US20210311762A1 | Guest cluster deployed as virtual extension of management cluster in a virtualized computing system | 30 |
| 9 | US20240129161A1 | Network segmentation for container orchestration platforms | 23 |
| 10 | US20230109368A1 | Autoscaling GPU applications in kubernetes based on GPU utilization | 19 |
Citation counts reflect influence within the searched corpus and skew toward older filings; a 2023 filing with fewer citations may still claim more current ground.
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 say about the field
Three findings that shape how a competitive or freedom-to-operate review of this space should be scoped.
The peak has already passed
Filings rose from zero in 2017 to a peak of 37 families in 2023, then eased. A midpoint reading of just 16 in 2022 confirms the climb was compressed into a few years rather than sustained — later entrants are filing into an already-dense claim landscape.
Compute-execution framing dominates
Almost every family in this set is classified under G06F, with H04L (43 records) layered on top. Very few claims lean on G06N (AI-based computing, 6 records), so AI-native scheduling logic remains a comparatively open drafting angle.
US prosecution sets the baseline
The United States receives more than four times the filings of the next-largest office (EPO, 15), with India (8), WIPO/PCT (6), China (4) and Austria (1) trailing. Any freedom-to-operate check that skips US prior art is checking the wrong corpus first.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to container orchestration and scheduling, with the prior art for and against each one.
Who is filing, and where momentum has stalled
Recent-year momentum across tracked assignees reads as flat or negative rather than accelerating — a signal that active filers have largely completed their core claim positions.
No assignee grew filings in the most recent year
Every assignee in the recent-momentum tracking — including VMware, Juniper Networks, Hewlett Packard Enterprise, DigitalOcean-scale players, IBM and SAP — shows zero filings in the latest tracked year, with Juniper Networks down 100% year-on-year. This is consistent with publication lag rather than an actual halt in R&D, but it also means the visible claim map for 2025-2026 is thin.
Co-filing is rare in this dataset
Only one co-assignee pairing rises above a single joint filing: IBM's parent entity alongside IBM Deutschland GmbH. Most families in this set are filed by a single assignee, suggesting orchestration and scheduling IP here is built in-house rather than through joint development agreements.
Older queue-management claims still anchor the field
The most-cited record in the set addresses queue management for cluster scheduling, followed closely by two container-orchestration filings addressing clustered/virtualized and decentralized deployment models. Newer entrants drafting around scheduling logic should check these first.
| Assignee | Recent year | YoY |
|---|---|---|
| VMware, Inc. | 0 | — |
| Juniper Networks, Inc. | 0 | -100% |
| Hewlett Packard Enterprise Development LP | 0 | — |
| DigitalOcean, Inc. | 0 | — |
| International Business Machines Corporation (IBM) | 0 | — |
| SAP SE | 0 | — |
| Oracle International Corporation | 0 | — |
| Microsoft Technology Licensing, LLC | 0 | — |
Where to take this analysis
A landscape count tells you where claims already sit. The next steps are about testing a specific filing or design against that map.
Run a freedom-to-operate check on a specific claim
Take a draft claim on autoscaling, bin packing or multi-tenant isolation and check it against the highest-cited US records first, since US filings make up two-thirds of this dataset.
Check claims in EurekaTrack assignee momentum before it shows up elsewhere
Zero recent-year filings across every tracked assignee likely reflects an 18-month publication lag rather than a real slowdown; monitor filings as they publish rather than relying on this year's count alone.
Set up monitoring in EurekaScope a white-space filing in AI-driven scheduling
With only 6 of 103 records touching G06N, an AI-model-based scheduling or interference-prediction claim has comparatively little prior art to design around today.
Explore white space in EurekaCommon questions on this patent landscape
The dataset tracks assignees including VMware, Juniper Networks, Hewlett Packard Enterprise, IBM and SAP, among others, ranked by patent family count in the accompanying table. Family counts are the fairer comparison than raw document counts because they remove the effect of continuation filings and multi-jurisdiction duplicates. No single assignee shows growth in the most recent tracked year, so current leadership reflects filings made earlier in the 2020-2023 window rather than fresh activity.
No — filings rose from zero in 2017 to a peak of 37 families in 2023, then flattened. The 2022 midpoint of 16 families sitting well below the 2023 peak shows the growth curve already crested rather than continuing to climb. Recent years show lower counts, but that is partly an artefact of the roughly 18-month lag between filing and publication, so 2025-2026 figures will rise somewhat as more records publish.
This dataset filters on IPC classes G06F9 (program execution and scheduling), H04L67 (network-application transmission protocols) and G06F11 (error detection and recovery), which together capture how orchestration claims are actually drafted. In practice, 93 of 103 records fall under the broader G06F subclass and 43 under H04L, confirming that scheduling patents are framed primarily as compute-execution claims with networking elements attached rather than as pure networking or AI patents.
AI-driven scheduling is comparatively open: only 6 of 103 records touch G06N (AI-based computing), versus 93 touching core compute execution. Cold-start latency mitigation, cross-cluster multi-tenancy isolation and edge/5G-native scheduling also show thinner claim density relative to the core autoscaling and bin-packing space. A first claim in these areas should tie a specific technical mechanism — such as a prediction model driving a scaling decision — to a measurable performance or isolation outcome, following the structure used in the SAP multi-tenancy interference filing.
US20230104787A1 covers autoscaling decisions driven specifically by a multi-tenant interference model that combines estimated performance, actual performance and a violation count per resource unit — it does not claim autoscaling generally. A scheduler that scales based on raw CPU or memory utilisation thresholds, without modelling tenant-to-tenant interference, sits outside this specific claim structure. Anyone building interference-aware autoscaling should review the claim language closely, since that is the narrower and more defensible ground the filing actually stakes out.
Research Container Orchestration and Scheduling in depth with Eureka
Go past this page: query the whole container orchestration and scheduling 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.