Hybrid Computing Data Processing via Message Passing Module
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hybrid computing environments face challenges in data processing due to the differences in architectures between host computers and accelerators, which hinder efficient data communication and task management across these disparate systems.
Innovation Solution
A system level message passing module (SLMPM) is used to manage threads of execution between host computers and accelerators, allowing for optimized data communication by selecting appropriate data communications modes and protocols, and abstracting the architectural differences to enable seamless interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If host computers and accelerators with different architectures are integrated to form a hybrid computing environment, then computational power and data processing efficiency are improved, but architectural differences create challenges in data communication and task management
Solution Approach 1:
The patent introduces a gateway as an intermediary component that mediates communication between the host computer and accelerator. The gateway translates data communication protocols and manages task coordination, allowing components with different architectures to interact seamlessly. This intermediary layer resolves the architectural incompatibility while maintaining high data processing efficiency.
Solution Approach 2:
The patent segments the hybrid computing system into distinct functional components: host computer, accelerator, and gateway. Each component operates independently with its own optimized architecture, and the gateway coordinates between them. This segmentation allows each component to be optimized for its specific function while reducing the overall integration complexity through modular design.
2Speed
If accelerators are optimized for speed of execution of specific computing functions, then processing speed is improved, but adaptability to different host computer architectures deteriorates
Solution Approach 1:
The gateway serves as an adapter that enables the accelerator to work with different host computer architectures. Instead of modifying the accelerator's core optimized architecture, the gateway handles protocol translation and interface adaptation, allowing the accelerator to maintain its high execution speed while becoming adaptable to various host systems.
Solution Approach 2:
The gateway provides universal interface capabilities that allow the accelerator to communicate with multiple types of host computers. The gateway implements multiple communication protocols and adaptation mechanisms, making the accelerator system universally compatible across different architectural platforms while preserving its specialized optimization for specific computing functions.
Data Source
AI summary
Data processing in a hybrid computing environment that includes a host computer having a host computer architecture; an accelerator having an accelerator architecture, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions; the host computer and the accelerator adapted to one another for data communications by a system level message passing module; and a host application process executing on the host computer. Data processing such a hybrid computing environment includes starting, at the behest of the host application process, a thread of execution on the accelerator; returning, by the system level message passing module to the host application process, a process identifier (‘PID’) for the thread of execution; and managing, by the host application process, the thread of execution on the accelerator as though the thread of execution were a thread of execution on the host computer.