/Patent Tracker
NvidiaPatent applicationAI & softwareInterest 3/10

Nvidia patents system to route data streams to specific software tasks

Patent drawing
Patent drawing from the official filing.

Nvidia describes a method for managing data flow in a cluster of computers. It uses agents to track data sources and route them to specific software services that need that information.

Why it mattersThis filing details internal infrastructure for organizing how data moves between different parts of a computing system.

In the filing’s own words

Approaches presented herein relate to the distribution and processing of stream data in a networked computing environment. A set of streaming data sources can be identified for a given cluster of nodes and the data from those data sources can be tracked, managed, and/or routed to a number of different microservices. Stream distribution and routing (SDR) agents may be used to manage a routing table that includes an identifier for each of the data sources. The individual agents may distribute information about the sensors to individual microservices, which may determine which data streams are relevant and then initiate individual processes and functions using the relevant data streams. Output results from the individual microservices may also be assigned an identifier, which may further be used by the individual agents to route outputs toward their particular microservices, which may then use the outputs from other microservices within their own workstreams.

What exactly they claim (claim 1)

1. A computer-implemented method, comprising: assigning an identifier to one or more data sources, the identifier of the one or more data sources being made available via a control bus; receiving a request to connect a data stream associated, with a data source of the one or more data sources with an available workload object of a plurality of workload objects that is communicatively coupled with the control bus; obtaining the identifier of the one or more data sources and an identifier associated with a particular workload object; populating a routing table with the identifier of the one or more data sources and the identifier associated with the particular workload object; performing a computing operation, using the data stream, by the particular workload object; and updating the routing table with a second identifier corresponding to a result generated by the particular workload object.

Facts

Company
Nvidia (Nvidia Corporation)
Official title
PROVISION AND CONFIGURATION OF VISIBILITY SERVICE IN CLUSTERS
Publication number
US20260329708A1
Application number
19723978
Filed
Jun 29, 2026
Published
Oct 22, 2026
Inventors
2
Claims
1

Classified under

G06F 11/2023 Physics & computingG06F 11/0709 Physics & computingG06F 11/2056 Physics & computingH04L 67/1097 Electricity & electronics

More from Nvidia in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.