US 12,271,636 B2
Analytics, algorithm architecture, and data processing system and method
Robert Bismuth, Seattle, WA (US); and Mike Stengle, Basel (CH)
Filed by Fermat International, Inc., Seattle, WA (US)
Filed on Nov. 6, 2023, as Appl. No. 18/502,275.
Application 18/502,275 is a continuation of application No. 18/075,576, filed on Dec. 6, 2022, granted, now 11,886,750.
Application 18/075,576 is a continuation of application No. 17/666,049, filed on Feb. 7, 2022, granted, now 11,550,512.
Application 17/666,049 is a continuation of application No. 16/729,727, filed on Dec. 30, 2019, granted, now 11,249,690.
Claims priority of provisional application 62/882,044, filed on Aug. 2, 2019.
Claims priority of provisional application 62/853,764, filed on May 29, 2019.
Claims priority of provisional application 62/801,839, filed on Feb. 6, 2019.
Prior Publication US 2024/0078049 A1, Mar. 7, 2024
Int. Cl. G06F 9/38 (2018.01); G06F 3/06 (2006.01)
CPC G06F 3/0661 (2013.01) [G06F 3/0613 (2013.01); G06F 3/067 (2013.01); G06F 9/3867 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A compute node for use in a data processing system, the compute node comprising:
a communications link communicatively coupling the compute node to one or both of a host compute system and an adjacent compute node in an execution pipeline;
a data store comprising records associated with a data processing operation;
a programmable logic component to execute the data processing operation in cooperation with the data store;
a node memory comprising data and instructions to support operation of the programmable logic component in connection with the data processing operation and to facilitate inter-node data communications with the adjacent node via the communications link, wherein the node memory further comprises algorithm memory including resource-specific memory locations to support the data processing operation;
a data mover component to facilitate intra-node data communications between the programmable logic component and the node memory; and
a storage interface component to facilitate intra-node data communications between the programmable logic component and the data store;
wherein the storage interface component utilizes a plurality of communications channels to transfer data between the programmable logic component and the data store; and
wherein the data processing operation is associated with a computationally expensive algorithm and the compute node is deployed in a distributed processing compute environment to execute the computationally expensive algorithm.