Knowledge Grid LLC is the current assignee of seven issued United States patents, all of which remain active. Six were granted between 2012 and 2015 and cover the foundations of a relational database engine: how data is organized and stored, how it is compressed, and how metadata about it is managed so that queries can be answered without scanning the underlying rows. The seventh, granted in 2022, covers the intelligent capture and fast transformation of granulated data summaries inside a database engine. Rough set theory — a formal mathematics for reasoning about data under uncertainty — is referenced across all seven documents, and approximate query processing runs throughout the most recent one. The earliest priority date is 2006, placing the underlying research almost two decades back, and further applications are pending. These patents describe the data-engine foundation rather than the whole of the platform built on it: the Temporal Data Grid extends that foundation into time-aware, entity-level knowledge structures developed since. Each patent number below links to its public record, where the claims, assignment history and full text can be read.
| Patent | Title | Granted | Priority | Status |
|---|---|---|---|---|
| US8266147 | Methods and systems for database organization | 2012-09-11 | 2006-09-18 | Active |
| US8417727 | System and method for storing data in a relational database | 2013-04-09 | 2010-06-14 | Active |
| US8521748 | System and method for managing metadata in a relational database | 2013-08-27 | 2010-06-14 | Active |
| US8700579 | Method and system for data compression in a relational database | 2014-04-15 | 2006-09-18 | Active |
| US8838593 | Method and system for storing, organizing and processing data in a relational database | 2014-09-16 | 2006-09-18 | Active |
| US8943100 | System and method for storing data in a relational database | 2015-01-27 | 2010-06-14 | Active |
| US11301467B2 | Systems and methods for intelligent capture and fast transformations of granulated data summaries in database engines | 2022-04-12 | 2018-06-29 | Active |
Assignee, status and dates are as recorded on Google Patents. The assignment history for each patent is visible on its linked record.
What the patents cover
The patents describe the data engine underneath the platform. They are the starting point, not the boundary of what Knowledge Grid has built since.
Storage, compression and metadata
Six patents granted 2012–2015 describe how a relational database organizes and compresses data, and how it keeps metadata about that data so questions can be answered from summaries instead of from raw rows.
Granulated summaries and approximate query processing
US11301467B2 (2022) covers capturing and transforming granulated data summaries inside a database engine — the mechanism behind answering queries approximately, at speed, without rescanning the underlying data.
Rough set mathematics
Rough set theory is referenced across all seven patents. It is the formal framework for reasoning about data under uncertainty that underpins the Temporal Data Grid and the Cognitive Data Layer built on it.
From the patented foundation to the Temporal Data Grid
The seven patents describe a database engine: how data is organized, compressed and summarized so that questions can be answered from granulated summaries rather than from raw rows. That engine is the foundation. The Temporal Data Grid is what Knowledge Grid built on top of it, and it goes further than the patented base in ways the patent documents do not describe — maintaining behavioral baselines per entity, preserving context across entities, events and time, and producing reusable knowledge structures that AI systems can query directly. Anyone reading the patents will find the mathematics and the storage engine; the time-aware layer, the entity modelling and the security-telemetry application are later work built on that base. Both statements are true at once: the foundation is patented, and the product has moved beyond what the foundation alone describes.