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 →Filing growth compares 2021 (45 records) with 2024 (33) — 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 529 records in scope (CR5), not by the ranked leaders only.
Resource isolation and overcommit sits at the intersection of scheduling, cache management and container runtime design: the mechanisms that let a host oversubscribe CPU, memory and last-level cache across tenants without one workload degrading another's latency. This landscape covers 529 published records filed or published between 2015 and mid-2026, drawn from claims combining resource isolation, performance interference or cache partitioning with implementation details like CPU throttling, memory pressure, overcommit ratios or container runtime behaviour. Publication lags filing by roughly 18 months, so the most recent one to two years in any trend chart will read lower than eventual filing volume once those applications publish.
The technology sits almost entirely inside electric digital data processing (G06F), with a substantial secondary presence in digital transmission (H04L) — consistent with isolation logic that has to coordinate across a network fabric, not just within a single host. Receiving-office data shows the United States as the dominant office by a wide margin, with the EPO, WIPO/PCT and Germany forming a secondary tier and India appearing as a notable filing venue.
Two views of the same 529-record dataset: how filing volume has moved year over year, and which IPC subclasses the claims fall under.
Annual filings climbed from 27 in 2017 to a peak of 71 in 2022. Using the last fully comparable years, filings fell 27% from 45 in 2021 to 33 in 2024 — a real cooling, though not a collapse. 2025 and 2026 figures are still incomplete due to publication lag and should not be read as a continuation of decline.
G06F (electric digital data processing) appears in 89.4% of the 529 records, confirming this is fundamentally a compute-scheduling and memory-management field. H04L (digital transmission) reaches 31.8%, reflecting isolation mechanisms that span networked, multi-host deployments. Smaller shares in G06N (AI-based computing, 4.2%), H04W (wireless, 4.2%) and G06T/G06V (image processing, under 4% combined) mark where isolation techniques are being adapted to adjacent workload types rather than where the core art lives.
Shares are the percentage of the 529 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
This page is one run against one query. Ask Eureka your own question about resource isolation and overcommit and every answer comes back with the patent numbers behind it.
Try EurekaA last-level cache controller uses a system state monitor and a cache partitioning module to combine a latency sensitivity factor, off-chip latency factors and per-core cache miss information into a weighted latency score. The controller aggregates these weighted latencies across cores and uses the result to decide whether a candidate cache-partition configuration should replace the current one.Assigned to MediaTek; published 2017-12-05.


| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20130185433A1 | Performance interference model for managing consolidated workloads in qos-aware clouds | 369 |
| 2 | US20250259085A1 | Convergent Intelligence Fabric for Multi-Domain Orchestration of Distributed Agents with Hierarchical Memory … | 208 |
| 3 | US20170264684A1 | Container runtime image management across the cloud | 187 |
| 4 | US20110225299A1 | Managing performance interference effects on cloud computing servers | 134 |
| 5 | US20210141655A1 | Unified resource management for containers and virtual machines | 97 |
| 6 | US20220100247A1 | Hierarchical power management apparatus and method | 77 |
| 7 | US20240007414A1 | Methods, systems, articles of manufacture and apparatus to optimize resources in edge networks | 72 |
| 8 | US20150058843A1 | Virtual hadoop manager | 72 |
| 9 | US20200257968A1 | Self-learning scheduler for application orchestration on shared compute cluster | 70 |
| 10 | US20240305533A1 | Radio resource planning and slice-aware scheduling for intelligent radio access network slicing | 67 |
Ranked by citation count within the searched corpus. Older records accumulate more citations by virtue of age, so treat this as a signal of influence on subsequent filings, not of current commercial relevance.
Each row carries its publication number; clicking a row searches Eureka by that number.
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 →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 →Four read-outs from the dataset that matter more than the raw record count.
The five most active filers account for 299 of the 529 records in scope, and the ranked ten reach 72.0%. New entrants are filing into a field where the dominant infrastructure and chip vendors have already staked broad claims around cache partitioning and workload interference.
Filings peaked at 71 in 2022. Comparing 2021 (45) to 2024 (33) — the last year with a complete publication picture — shows a 27% decline. Several previously active filers show 0 filings in the latest tracked year, though that may partly reflect publication lag rather than an exit.
Nearly nine in ten records carry a G06F classification, confirming the core art is scheduling and memory management. Roughly a third also carry H04L, indicating that isolation logic increasingly has to account for network-distributed, multi-host deployments rather than a single machine.
The most-cited record in scope addresses performance interference modelling for consolidated workloads in QoS-aware clouds — an early framing that later container-runtime and cache-partitioning claims build on. High citation counts here reflect age and foundational status more than current filing activity.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to resource isolation and overcommit, with the prior art for and against each one.
Filing activity is led by infrastructure and chip vendors with deep runtime and hardware stacks, alongside cloud and telecom entrants building isolation into orchestration platforms.
The leading assignee's 130 records dwarf the fifth-place count of 25, indicating a filer that has treated resource isolation as core IP rather than an incidental claim area within broader chip or platform patents.
The drop from the leader's 130 to fifth place's 25 is steep, and the gap from fifth to tenth (11) is comparatively shallow — suggesting a compact group of serious filers below the leader, followed by many single- or few-filing entrants.
Only ten co-assignee pairs appear in the dataset, and the strongest link connects a parent company to its own regional development subsidiary rather than an external partnership — cross-company joint filing is not yet a norm in this space.
| Assignee | Recent year | YoY |
|---|---|---|
| UnifabriX Ltd. | 2 | -91% |
| LG Electronics Inc. | 1 | — |
| Intel Corp. | 0 | -100% |
| VMware Inc. | 0 | -100% |
| Microsoft Technology Licensing, LLC | 0 | -100% |
| Oracle International Corp. | 0 | — |
| International Business Machines Corporation | 0 | — |
| ROBIN SYSTEMS INC | 0 | — |
The dataset points to a field with a clear leadership tier and a compute-heavy claim base, but the sub-branches worth watching depend on freedom-to-operate specifics no landscape summary can settle on its own.
With 56.5% of records held by five filers, a new filing in cache partitioning or overcommit control is likely to sit close to existing claims. A targeted search against the leader's portfolio is a reasonable first step before drafting.
Run a freedom-to-operate search in EurekaSeveral major filers show sharp year-over-year drops that may reflect publication lag rather than a genuine pullback. Re-checking filing momentum in six to twelve months will separate a real slowdown from a reporting artifact.
Set up a monitoring alert in EurekaOne assignee leads clearly with 130 records, well ahead of fifth place at 25 and tenth place at 11. The top five filers combined hold 299 of the 529 records in scope, or 56.5%, and the top ten reach 72.0%. This is a field where a handful of infrastructure and chip vendors have built deep portfolios rather than one dominated by a single company alone.
Filing activity rose from 27 records in 2017 to a peak of 71 in 2022, then eased: comparing 2021 (45) to 2024 (33), the last fully comparable years, shows a 27% decline. Figures for 2025 and 2026 look lower still, but that is expected because publication typically lags filing by about 18 months, so those years are not yet complete. Treat the 2022 peak and the 2021-2024 decline as the reliable read, not the most recent bars on the chart.
US9836396B2, assigned to MediaTek, claims a last-level cache controller that computes a weighted latency score per processor core from a latency sensitivity factor, off-chip latency factors and cache miss information, then aggregates those scores to decide whether to switch to a new cache-partition configuration. It is a specific control-logic approach to last-level cache partitioning, not a claim over cache partitioning generally. Anyone building a competing cache-partition controller should check whether their decision logic relies on a similarly weighted, per-core aggregated latency score before assuming clearance.
Electric digital data processing (G06F) appears in 89.4% of the 529 records in scope, making it by far the dominant classification, consistent with a field centred on scheduling, memory management and cache control logic. Digital transmission (H04L) is the strongest secondary class at 31.8%, reflecting isolation mechanisms that operate across networked, multi-host environments. Classes tied to AI computing, wireless networks and image processing each sit under 5%, marking adjacent application areas rather than the core art.
The smaller IPC classes point to under-claimed intersections: cache partitioning applied to AI/ML accelerator workloads, overcommit ratio control for latency-sensitive video pipelines, and isolation logic for vehicle compute platforms (B60R) all show low record counts relative to the 89.4% concentration in core G06F claims. Co-filing is also rare, with only 10 co-assignee pairs identified and most of those intra-corporate, suggesting few defensive cross-licensing arrangements yet exist in these adjacent branches. A first mover with a well-drafted claim in one of these narrower intersections would face less prior art than in core cache-partitioning or CPU-throttling claims.
Go past this page: query the whole resource isolation and overcommit 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.
Help us improve
Something wrong, out of date, or missing on this page? Tell us and it goes straight to the team that maintains it.
This page