US 11,693,662 B2Granted

Method and apparatus for configuring a reduced instruction set computer processor architecture to execute a fully homomorphic encryption algorithm

AssigneeCornami, Inc.

InventorsMorris Jacob Creeger, Tianfang Liu, Frederick Furtek, Paul L. Master

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods for configuring a reduced instruction set computer processor architecture to execute fully homomorphic encryption (FHE) logic gates as a streaming topology. The method includes parsing sequential FHE logic gate code, transforming the FHE logic gate code into a set of code modules that each have in input and an output that is a function of the input and which do not pass control to other functions, creating a node wrapper around each code module, configuring at least one of the primary processing cores to implement the logic element equivalents of each element in a manner which operates in a streaming mode wherein data streams out of corresponding arithmetic logic units into the main memory and other ones of the plurality arithmetic logic units.

Claims8

Dependent claims

A method for configuring a reduced instruction set computer processor architecture to process a Discrete Fourier Transform (DFT) of a finite-length sequence N, wherein the computer processor architecture includes a plurality of primary processing cores, each primary processing core comprising a local memory, and a plurality of arithmetic logic units, each primary processing core having an associated node wrapper, the associated node wrapper including access memory associated with each arithmetic logic unit, a load/unload matrix associated with each arithmetic logic unit, the method comprising:

(a) applying a Decimation-in-Frequency algorithm to the DFT to decompose the DFT of a finite-length sequence N into two derived DFTs each of a length N/2;
(b) constructing a logic element equivalent of each stage of the derived DFTs in which inputs and outputs are composed of real and imaginary components;
(c) repeating (a) and (b) for each stage of the DFT except for the endpoint stages of the DFT, wherein the endpoint stages are a first stage of the DFT and a last stage of the DFT;
(d) for each stage of the DFT configuring the logic elements to provide all of the required inputs and outputs of the DFT and, for each endpoint, configuring a logic element equivalent of the corresponding stage of the derived DFTs in which inputs to the first stage and outputs of the last stage are composed of only real components;
(e) configuring at least one primary processing core of the computer processor architecture to implement the logic element equivalents of each stage of the derived DFTs in a manner which operates in a streaming mode wherein data streams out of corresponding arithmetic logic units into the local memory and other ones of the plurality arithmetic logic units;
and
(f) configuring the computer processor architecture to couple the output of each stage on the DFT to the input of a subsequent stage.

4 more elements

A computer processor architecture for processing a Discrete Fourier Transform (DFT) of a finite-length sequence N, the computer processor architecture comprising:

a plurality of primary processing cores, each primary processing core comprising a local memory, and a plurality of arithmetic logic units, each primary processing core having an associated node wrapper, the associated node wrapper including access memory associated with each arithmetic logic unit, a load/unload matrix associated with each arithmetic logic unit;
wherein multiple stages of the DFT are constructed by, (a) applying a Decimation-in-Frequency algorithm to the DFT to decompose the DFT of a finite-length sequence N into two derived DFTs each of a length N/2, (b) constructing a logic element equivalent of each stage of the derived DFTs in which inputs and outputs are composed of real and imaginary components, (c) repeating (a) and (b) for each stage of the DFT except for the endpoint stages of the DFT, wherein the endpoint stages are a first stage of the DFT and a last stage of the DFT, (d) for each stage of the DFT configuring the logic elements to provide all of the required inputs and outputs of the DFT and, for each endpoint, configuring a logic element equivalent of the corresponding stage of the derived DFTs in which inputs to the first stage and outputs of the last stage are composed of only real components;
wherein at least one primary processing core of the computer processor architecture is configured to implement the logic element equivalents of each stage of the derived DFTs in a manner which operates in a streaming mode wherein data streams out of corresponding arithmetic logic units into local memory and other ones of the plurality arithmetic logic units;
and
wherein the computer processor architecture is configured to couple the output of each stage on the DFT to the input of a subsequent stage.

2 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/743,257Filed 15 Jan 2020 · continuation in part
Priority: 4 May 2018
Published: US 2020/0213079 A12 Jul 2020
Granted: 4 Jul 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification10 codes

G06F 7/544
Using non-contact-making devices, e.g. tube, solid state device; using unspecified devices: for evaluating functions by calculation
G06F 9/3812
Instruction prefetching: with instruction modification, e.g. store into instruction stream
G06F 7/4806
Using non-contact-making devices, e.g. tube, solid state device; using unspecified devices: computations with complex numbers
G06F 7/57
Using non-contact-making devices, e.g. tube, solid state device; using unspecified devices: arithmetic logic units [ALU], i.e. arrangements or…

Family

Application15/970,915Parent of this continuation-in-part · filed 4 May 2018
Application62/883,967Us provisional application · filed 7 Aug 2019
US 2020/0213079 A1Pre-grant publication · published 2 Jul 2020
US 11,294,851 B2Parent of this continuation-in-part
US 2019/0340152 A1Published application
US 11,693,662 B2This patent

From the US filing history. Foreign family members aren’t listed.