Software Parallelization Scoring for Hardware Architecture Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for determining the most suitable hardware architecture for software applications are tedious, expensive, and require detailed knowledge of the target architecture and programming language, necessitating multiple trials and code adaptation.

Innovation Solution

A method that determines a hardware architecture by executing a software application with varying input data sets, calculating a parallelization score based on output changes, and selecting the architecture using a table associating score ranges with hardware types, without needing to know the hardware or programming language specifics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If dynamic program analysis is used to determine suitable hardware architecture, then accurate evaluation of application performance is achieved, but the process becomes tedious and expensive requiring detailed knowledge of target architecture and programming languages

Engineering Contradiction:

Improveapplication performance evaluation accuracyVSAvoidmethod complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary metric calculation system that automatically evaluates application characteristics without requiring expert knowledge. The system uses intermediate metrics (execution time, resource consumption, parallelization potential) as mediators between the application code and hardware architecture selection, replacing the need for dynamic program analysis by human experts.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The method enables self-service by allowing the system to automatically determine suitable hardware architecture through automated metric calculation and evaluation. The application itself provides the necessary information through its execution characteristics, eliminating the need for external expert analysis and making the process self-sufficient.

Inventive Principle:

Principle #25Self-service

2Reliability

If multiple execution tests are performed on different target hardware architectures, then the most suitable architecture is identified, but the process becomes expensive and time-consuming

Engineering Contradiction:

Improvehardware architecture selection accuracyVSAvoidarchitecture determination time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by calculating key metrics (execution time, resource consumption, parallelization potential) in advance during a single test execution. These pre-calculated metrics are then used to evaluate multiple hardware architectures without requiring actual execution tests on each target platform, significantly reducing time and cost while maintaining selection accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The method creates a virtual model or copy of the evaluation process by using metric-based simulation instead of physical execution tests. The system copies the essential characteristics of multiple hardware architectures and evaluates the application against these models through calculated metrics, avoiding the need for actual multiple execution tests.

Inventive Principle:

Principle #26Copying

3Productivity

If application code is analyzed in detail to evaluate parallel operation possibilities, then optimal parallelization is achieved, but the process requires detailed knowledge of the application structure and becomes complex

Engineering Contradiction:

Improveparallel operation efficiencyVSAvoidcode analysis complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces manual code analysis (mechanical process requiring human expertise) with an automated metric calculation system. The system substitutes the mechanical process of detailed code inspection with automated calculation of parallelization potential metrics, maintaining the ability to evaluate parallel operation possibilities while eliminating the need for detailed code analysis by humans.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4560459B1Method for selecting a hardware architecture for implementing a software application

Publication Date: 2026.04.29 CENT NAT DETUD SPATIALES (CNES)

Figure 1~2
Figure 3~4

AI summary

The invention relates to a method for determining a target hardware architecture suitable for a software application, the method comprising steps consisting in: defining a first input data set (ID0) of the application (AL2); executing the application by a processor (PRC) from the first input set to obtain a first output data set (OD0); executing several iterations each comprising: selecting an input data item; generating a second input set (ID1, ID2) from the first input set, by modifying the selected input data item; executing the application from the second input set to obtain a second output set (OD1, OD2); and calculating a metric relating to the first and second input and output sets of the iteration; calculating a parallelization score (PS) from the metrics calculated at each iteration; and determining the target architecture according to the parallelization score.