Cryptographic Batch Processing via Mutualized Calculations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Cryptographic computations are resource-intensive and can exceed the computational capacity of systems, particularly when multiple processes or devices request operations simultaneously, leading to increased execution time or the need to reject requests.

Innovation Solution

A method that involves a processing system waiting for a predetermined time to collect multiple requests for cryptographic operations before performing mutualized calculations, which reduces resource consumption and execution time by executing shared calculations only once, rather than sequentially for each operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If cryptographic operations are performed sequentially for each request, then execution time for each operation is minimized, but global execution time increases and resource consumption increases

Engineering Contradiction:
Improveglobal execution timeVSAvoidthroughput
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The patent combines multiple cryptographic operations into a single batch processing operation. Instead of executing each cryptographic operation separately and sequentially, the system collects multiple requests and performs them simultaneously using mutualized calculations, thereby reducing global execution time and improving throughput.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system performs preliminary actions by collecting and queuing multiple cryptographic requests before executing them. A request queue is maintained, and operations are gathered in advance to enable batch processing, which optimizes resource utilization and reduces overall execution time.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If multiple cryptographic operations are handled concurrently, then throughput increases, but execution time for each operation increases due to context saving/restoring overhead

Engineering Contradiction:
ImprovethroughputVSAvoidexecution time per operation
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

Multiple cryptographic operations are merged into a single batch processing operation. By combining operations that would otherwise require separate context switches and execution threads, the system eliminates repeated context saving and restoring overhead, thereby reducing execution time per operation while maintaining high throughput.

Inventive Principle:
Principle #5Merging (Combining)

3Use of energy by moving object

If a waiting time is introduced to collect multiple requests, then resource consumption decreases, but execution time increases due to idle waiting period

Engineering Contradiction:
Improveresource consumptionVSAvoidtotal execution time
Core Design Contradiction:
Use of energy by moving objectVSLoss of time

Solution Approach 1:

The system introduces a waiting period that serves as a preliminary action to collect multiple cryptographic requests before executing them in a batch. This waiting time is intentional and productive, as it gathers multiple operations that can then be executed efficiently together, reducing overall resource consumption despite the apparent idle period.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The waiting time is made dynamic rather than fixed. The system adjusts the waiting period based on the arrival rate of requests and the complexity of operations, optimizing the balance between collecting sufficient requests for efficient batch processing and minimizing unnecessary waiting time that would increase total execution time.

Inventive Principle:
Principle #15Dynamics

4Use of energy by moving object

If cryptographic operations are performed in parallel on single CPU, then resource utilization increases, but system complexity increases due to scheduling overhead

Engineering Contradiction:
ImproveCPU utilizationVSAvoidscheduling complexity
Core Design Contradiction:
Use of energy by moving objectVSDevice complexity

Solution Approach 1:

The patent segments the scheduling complexity by introducing a simple request queue mechanism that separates request collection from execution. This segmentation allows the system to handle multiple operations without complex real-time scheduling, as requests are simply queued and processed in batches when ready, thereby increasing CPU utilization while keeping scheduling complexity manageable.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS12177347B2Method for performing mutualized calculations of multiple cryptographic operations
Publication Date: 2024.12.24 THALES DIS FRANCE SA
  • US12177347B2 patent drawing
  • US12177347B2 patent drawing

AI summary

Provided is a method for performing a plurality of cryptographic operations, that upon reception of a request to perform one of said cryptographic operations, prevents an execution by said processing system of said requested cryptographic operation until a predetermined waiting time (G) has elapsed, and before said predetermined waiting time has elapsed, receives one or more requests to perform another cryptographic operation, and after said predetermined waiting time (G) has elapsed, answers (S3) said requests by executing operations comprising mutualized calculations. The method determines said waiting time depending on execution times of said cryptographic operations to be performed and of said mutualized calculations.