US 11886750

FOR IMMEDIATE RELEASE

Patent Analysis: U.S. Patent No. 11,886,750

Date: April 26, 2026

Overview

This report provides a summary of United States Patent number 11,886,750, titled "Analytics, algorithm architecture, and data processing system and method." The patent, issued on January 30, 2024, is assigned to Fermat International Inc. The inventors are listed as Robert Bismuth and Mike Stengle. The application for this patent was filed on December 6, 2022.

Abstract

The patent describes a system and method that utilizes a distributed hardware architecture for data processing and analytics. This system can operate independently or in conjunction with a specific data structure. A key feature is the use of a "compute node" that can manage and execute data processing tasks independently of a host computer system. The patent also covers a unique algorithm architecture and processing system.

Independent Claims in Plain Language

An independent claim represents the broadest definition of the invention. U.S. Patent 11,886,750 contains three independent claims:

• Claim 1: This claim describes a method for executing data processing operations. It involves a "compute node" connected to a host computer but capable of working independently. This node contains a programmable logic component (like an FPGA) and memory. A key part of the method is a set of instructions that allow the logic component to reformat data. Specifically, it can rearrange data from a structure where each record has multiple types of information (fields) into a new structure where each record contains only a single type of information. The data is then transferred between the logic component and memory using multiple communication channels.

• Claim 11: This claim details a data processing system that works with a host computer. The system includes a "router module" that connects to the host and a "compute node." This compute node has its own processor, memory, and data storage. The programmable logic within the node is designed to reformat data in the same way as described in Claim 1: changing data from a record-based structure to a field-based structure. This reformatted data is then transferred between the logic and storage using multiple communication channels.

• Claim 20: This claim describes a more complex data processing system. It features a "management node" connected to a host computer and a "memory-supported compute node." This is then connected to a series of additional compute nodes, forming a "pipeline." The first compute node in this chain has its own memory and storage. It receives a "bit stream" of instructions for an algorithm and starts the process by using data from its storage. The results are then passed to the next node in the pipeline. Each subsequent node in the pipeline then performs its part of the algorithm using the results from the previous node. This allows for a chain of processing, where each node can specialize in a part of the overall task.

Litigation Status

A search of the U.S. Court of Appeals for the Federal Circuit (CAFC) dockets for the year 2026 has not revealed any litigation involving U.S. Patent No. 11,886,750. However, this does not preclude the possibility of pending or future litigation in other courts.

Disclaimer: This summary is for informational purposes only and does not constitute legal advice. The plain-language descriptions of the claims are intended to provide a general understanding and are not a substitute for a detailed legal analysis of the patent.