Intermediary Workflow Representation for Computational Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of using computational devices for workflows involves coordinating multiple technologies across domains, leading to significant investments of time and resources, and migrating workflows between devices requires rewriting code, which is inefficient and non-portable.
Innovation Solution
The implementation of an intermediary format representation of computational workflows, which includes generic instructions that can be executed across various computational devices, reducing the need for multiple technology implementations and facilitating interoperability by converting input representations into a standardized format that can be executed on different devices with minimal changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple technology implementations are used to support different computational devices, then device compatibility is improved, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary representation format that serves as a mediator between diverse computational devices and their respective technology implementations. This intermediary format allows workflows to be represented in a device-agnostic manner, enabling compatibility across different devices without requiring each device to implement multiple technology stacks. The intermediary representation decouples the workflow logic from device-specific implementations, thereby reducing system complexity while maintaining broad device compatibility.
Solution Approach 2:
The intermediary representation format is designed to be universal, capable of representing workflows for multiple types of computational devices through a single standardized structure. This universal format can encode various computational operations and data structures in a way that is interpretable by different device types, eliminating the need for separate technology implementations for each device and reducing overall system complexity.
2Adaptability or versatility
If code is rewritten to migrate workflows between devices, then workflow portability is improved, but development time and resources increase
Solution Approach 1:
The patent enables workflow copying by representing workflows in an intermediary format that can be replicated and transferred between devices without modification. The intermediary representation captures the essential logic and structure of workflows in a device-independent manner, allowing them to be copied from one computational device to another while maintaining functionality, thereby eliminating the need for time-consuming code rewriting and reducing development time.
Solution Approach 2:
The intermediary representation acts as a mediator that preserves workflow integrity during migration between devices. By translating workflows into this standardized format, the system enables seamless portability without requiring developers to rewrite code for each target device, thus maintaining workflow portability while minimizing the time and resources invested in migration activities.
3Productivity
If device-specific formats are used for computational workflows, then execution efficiency is improved, but interoperability between devices deteriorates
Solution Approach 1:
The patent segments the workflow representation into two distinct layers: an intermediary representation layer that ensures interoperability across devices, and device-specific execution layers that optimize for individual device characteristics. The intermediary format handles the universal aspects of workflow logic, while device-specific implementations handle the execution optimizations. This segmentation allows the system to maintain both interoperability through the common intermediary format and execution efficiency through tailored device implementations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method may include receiving, by at least one processing circuit, an input representation (332) of a computational workflow, wherein the input representation (332) may include at least one instruction in an input format, and generating, by the at least one processing circuit, based on the input representation (332), an intermediary format representation (334) of the computational workflow, wherein the intermediary format representation (334) may include at least one intermediary format instruction for a computational device. The at least one intermediary format instruction may include an instruction to perform, by the computational device, at least one of a load operation, a store operation, or a computational operation. The generating the intermediary format representation (334) may include generating the intermediary format representation (334) based on an arrangement of the computational workflow.