Server GPGPU Offload for Mobile Computing Bottlenecks

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Mobile devices face challenges in executing programs using general-purpose graphics processing units (GPGPU) due to limited power and computing speed, leading to increased manufacturing costs and reduced efficiency when trying to offload specific computations to a server.

Innovation Solution

A computing method and system where a client sends program code and resource data to a server equipped with a GPGPU for execution, allowing the server to process the data and send results back to the client, while the client can also perform some computations internally to optimize processing efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Power

If a server is programmed for specific purpose to execute GPGPU programs, then the server can provide dedicated computing capability, but the adaptability to different computation types is reduced

Engineering Contradiction:

Improvecomputing capabilityVSAvoidadaptability to different computation types

Core Design Contradiction:

PowerVSAdaptability or versatility

Solution Approach 1:

The server is designed with a GPGPU that can execute various types of computation programs through a unified programming interface. The system accepts programs in different formats (CUDA, OpenCL, HIP) and translates them to a common intermediate representation, enabling the same hardware to handle diverse computational workloads without requiring separate dedicated hardware for each computation type.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Speed

If GPGPU is mounted in mobile device to execute programs locally, then computing speed is improved, but manufacturing cost increases

Engineering Contradiction:

Improvecomputing speedVSAvoidmanufacturing cost

Core Design Contradiction:

SpeedVSEase of manufacture

Solution Approach 1:

The patent extracts the GPGPU computing capability from the mobile device and relocates it to a remote server. The mobile device only retains the client software and communication interfaces, while the expensive GPGPU hardware is situated in the cloud. This allows mobile devices to access high-performance computing through network connections without physically containing the costly hardware, thereby reducing manufacturing costs while maintaining computing speed.

Inventive Principle:

Principle #2Taking out (Extraction)

3Ease of operation

If mobile device executes GPGPU program locally, then processing is performed on-device, but computational burden on device increases

Engineering Contradiction:

Improveprocessing locationVSAvoidcomputational burden

Core Design Contradiction:

Ease of operationVSPower

Solution Approach 1:

The patent introduces a cloud server as an intermediary between the mobile device and the GPGPU computation. The mobile device sends computation tasks to the server, which executes them on the GPGPU and returns results. This intermediary architecture shifts the computational burden from the mobile device to the server, allowing the device to maintain ease of operation while reducing its power consumption and processing requirements.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS9467532B2Server, arithmatic processing method, and arithmatic processing system

Publication Date: 2016.10.11 IND ACADEMIC COOP FOUND YONSEI UNIV

AI summary

A computing method is provided which includes calling a general purpose graphics processing subroutine for execution of a target program by a client; sending a program code and resource data for execution of the target program to a server by the client; and executing the program code using a general purpose graphics processing unit by the server.