/Patent Tracker
GooglePatent applicationAI & softwareInterest 4/10

Google patents method to place data near where it is used in cloud systems

Google describes a system that looks at past computer queries to see which pieces of data are often used together. It then uses a mathematical model to decide where to copy that data in a global network, ensuring it is stored close to where it is needed for faster access.

Why it mattersThis improves the speed of large-scale data analysis by reducing the time computers spend waiting for data to travel across the network.

In the filing’s own words

Historical analysis of query patterns is used to discover relationships between data sets. These relationships are used to make optimal decisions about where to place data in a globally distributed environment in which locality of data is an important factor in providing good query performance. A mixed integer programming model is used to solve a constraint based system which balances the need to have data kept local with other data and the cost of replicating data across a low-bandwidth network.

What exactly they claim (claim 1)

1. A method for generating assignments of replicated data in a distributed system, comprising: storing a first set of data sets owned by a first organization on a first set of storage devices in the distributed system; analyzing query logs having transactions between the first organization and a second organization to identify sets of linked data that indicates one or more transactions between the first organization and a second organization, wherein the sets of linked data are identified based on queries that indicate data of the sets of linked data are joined with the first set of data sets; generating an assignment map based on a model, the assignment map indicating where to store the sets of linked data; and replicating storage of the sets of linked data on the first set of storage devices in the distributed system based on the assignment map.

Facts

Company
Google (Google LLC)
Official title
System for Optimizing Storage Replication in a Distributed Data Analysis System Using Historical Data Access Patterns
Publication number
US20260267874A1
Application number
19664558
Filed
Apr 30, 2026
Published
Sep 10, 2026
Inventors
4
Claims
20

Classified under

G06F 16/27 Physics & computingG06F 16/219 Physics & computingG06F 16/2358 Physics & computingG06F 16/285 Physics & computingG06F 17/10 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google 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.