Storage Controller Selective Encryption for Computation and Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Large-scale data operations between a central processing unit (CPU) and storage devices are hindered by inefficient data transfer speeds, particularly when using accelerator circuits like GPUs and FPGAs for AI or machine learning computations, and there is a need for enhanced data security through encryption.

Innovation Solution

The storage device selectively encrypts data using either a homomorphic or symmetric/asymmetric encryption algorithm based on operation mode, determined by command fields indicating computation and encryption status, and stores the encrypted data in non-volatile memory.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is encrypted using traditional encryption algorithms, then data security is improved, but computation efficiency deteriorates

Engineering Contradiction:
Improvedata securityVSAvoidcomputation efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent changes the encryption parameter (algorithm type) based on the operation mode. When computation is needed, homomorphic encryption is used; when only storage is needed, symmetric encryption is used. This dynamic parameter selection resolves the contradiction between security and computation efficiency.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system dynamically selects encryption algorithms based on operation modes determined by command fields. The encryption circuit switches between homomorphic encryption mode and symmetric encryption mode depending on whether computation or storage is the primary function, making the system adaptive rather than static.

Inventive Principle:
Principle #15Dynamics

2Reliability

If encryption algorithms are selected based on operation mode, then data security is improved, but device complexity increases

Engineering Contradiction:
Improvedata securityVSAvoidencryption circuit complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The encryption circuit is segmented into functionally distinct units: a homomorphic encryption circuit for computation modes and a symmetric encryption circuit for storage modes. This segmentation allows each unit to be optimized for its specific function while managing overall system complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The encryption circuit is designed as a multi-functional component that can perform both homomorphic encryption and symmetric encryption based on operation mode selection. This universal design consolidates functionality into a single circuit structure, reducing complexity compared to having separate dedicated circuits for each encryption type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If data transfer speed is increased for AI computations, then productivity is improved, but data security deteriorates

Engineering Contradiction:
Improvedata transfer speedVSAvoiddata security
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

Data is encrypted before being transferred to or stored in the storage device. By performing encryption as a preliminary action, the data remains secure throughout the transfer and storage process, eliminating the need to choose between speed and security after the fact.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The encryption circuit acts as an intermediary between the host and the storage device, ensuring that data is protected during transfer without becoming a bottleneck. The intermediary encrypts data efficiently using appropriate algorithms based on operation mode, maintaining both security and transfer performance.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS12406100B2Storage device including storage controller and operating method
Publication Date: 2025.09.02 SAMSUNG ELECTRONICS CO LTD
  • US12406100B2 patent drawing
  • US12406100B2 patent drawing
  • US12406100B2 patent drawing

AI summary

An operating method for a storage device including a storage controller and a non-volatile memory may include; receiving a command including data and a field related to the data from a host, determining an operation mode based on the command, selectively encrypting the data based on the operation mode to generate selectively encrypted data, and storing the selectively encrypted data in the non-volatile memory, wherein the selectively encrypting of the data includes encrypting the data based on a first encryption algorithm when the operation mode is a first operation mode, and encrypting the data based on a second encryption algorithm different from the first encryption algorithm when the operation mode is a second operation mode.