US 8190610

The information provided is current as of April 26, 2026.

Here is a concise summary of US Patent 8190610:

Title: MapReduce for distributed database processing

Current Assignee: Starboard Value Intermediate Fund LP, as Collateral Agent. Note: The Google Patents page states that the listed assignees may be inaccurate and that Google has not performed a legal analysis to verify accuracy. The "Legal Events" section of the provided document indicates an assignment to "STARBOARD VALUE INTERMEDIATE FUND LP, AS COLLATERAL AGENT" on March 29, 2021.

Inventors: Ali Dasdan, Hung-Chih Yang, Ruey-Lung Hsiao

Filing Date: 2006-10-05

Issue Date: 2012-05-29

Abstract: The patent enhances the MapReduce programming methodology by treating an input data set as multiple grouped sets of key/value pairs. This allows map processing to occur independently on two or more related, potentially heterogeneous datasets (e.g., sharing a common primary key). The intermediate key/value pair results for a specific key can then be processed together in a single reduce function by using a different iterator for intermediate values from each group. These iterators can be arranged within reduce functions as desired.

Plain-Language Overview of Independent Claims:

• Claim 1 (Method of Data Processing): This claim describes a method for processing data in a distributed system, where the data is organized into multiple "data groups." A key aspect is that at least two of these groups have different schemas (different structures) but share a common key. The method involves:

  • Mapping: Data from each group is divided into partitions, and each partition is sent to a mapping function. These mapping functions are customized by the user to process the data independently for each group. The mapping produces "intermediate data" that is specific to its original data group, with different sets of values generated for different data groups. The different schemas and their corresponding intermediate data still share a common key.
  • Reducing: The intermediate data from all groups is then processed together. The processing for each group's intermediate data is tailored to that specific group, ultimately merging the different intermediate data based on the shared common key to produce at least one output data group.
  • The entire mapping and reducing process is performed by a distributed system.

• Claim 17 (Computer System for Data Processing): This claim describes a computer system with at least one processor and memory that is configured to perform the data processing method outlined in Claim 1. It emphasizes the system's capability to execute the partitioning, mapping, and reducing steps, specifically handling heterogeneous data groups and merging their intermediate data based on a common key in a distributed environment.

• Claim 33 (Map-Reduce Method for Heterogeneous Groups): This claim details a MapReduce method specifically designed for processing data from multiple groups that possess "different schemas." The method involves:

  • First Data Set Processing: A first data set, part of a first data group with a first schema and key-value pairs, is partitioned and processed by mapping functions to create a first intermediate data set.
  • Second Data Set Processing: Similarly, a second data set from a second data group, with a distinct second schema and key-value pairs, is partitioned and processed by mapping functions to yield a second intermediate data set. This second intermediate data set will differ from the first.
  • Crucially, despite the differing schemas, both the original schemas and the resulting intermediate key-value pairs share a "key in common."
  • Reducing Together: Both the first and second intermediate data sets are then combined (reduced) to produce an output data set. This reduction is achieved by "iterating" over at least one common key present in both intermediate data sets. The resulting output data set will have a schema different from the initial first and second schemas.
  • These mapping and reducing operations are carried out by a distributed system.

• Claim 40 (Computer System for Heterogeneous Group Map-Reduce): This claim describes a computer system with at least one processor and memory configured to execute the MapReduce method specified in Claim 33. It highlights the system's capability to handle the mapping and reduction of data from distinct groups with different schemas but a common key, ultimately producing a merged output data set through iteration on that common key.

USPTO and CAFC 2026 Dockets:
The provided patent text, fetched on 2026-05-29, already contains recent legal status information. It states the patent is "Active" and expires on 2029-10-14. It also lists several litigation events:

• Multiple US district court cases filed in Texas Eastern, Texas Western, Texas Northern, and Nevada District Courts, dating from 2021 to 2023.
• Multiple PTAB (Patent Trial and Appeal Board) cases, including IPR2024-00659 (filed, Not Instituted - Merits), IPR2024-00303 (filed, Not Instituted - Merits), IPR2023-00689 (Settlement), and IPR2022-01405 (Settlement).

A search for "CAFC dockets US8190610 2026" did not immediately return specific CAFC docket numbers related to US8190610 in 2026. The Federal Circuit website details how to find case records, but direct docket information for a specific patent number for the current year (April 2026) was not presented in the search results. Given that the provided Google Patents data (fetched 2026-05-29) does not explicitly mention CAFC cases for this patent family beyond district court and PTAB proceedings (where appeals to CAFC would typically originate), and no specific CAFC dockets for 2026 were found in the search results, there is no authoritative information at this time to report on active CAFC dockets for US8190610 specifically in 2026. Appeals from PTAB decisions or district court judgments could be ongoing or initiated in 2026, but specific CAFC docket numbers are not readily available from the provided data or general search.