Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A GPU task scheduling system and method based on kernel merging

A task scheduling and task technology, applied in multi-programming devices, program control design, instruments, etc.

Active Publication Date: 2020-10-27
XI AN JIAOTONG UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, CUDA can only be used in heterogeneous systems with NVIDIA GPUs as acceleration devices. Therefore, at the end of 2008, OpenCL, a cross-platform heterogeneous parallel programming framework standard jointly formulated by many companies, appeared, which is applicable to any parallel system.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A GPU task scheduling system and method based on kernel merging
  • A GPU task scheduling system and method based on kernel merging
  • A GPU task scheduling system and method based on kernel merging

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057] Embodiments of the present invention are described below through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various modifications or changes can be made to the details of this specification based on different viewpoints and applications without departing from the spirit of the present invention.

[0058] The purpose of this embodiment is to provide a GPU task scheduling method and system based on Kernel merging, which is used to solve the problems of low utilization rate of GPU scheduling resources and low throughput rate in the prior art. The principle and implementation of a kind of GPU task scheduling method and system based on Kernel merging of the present invention will be described in detail below, so that those skilled in the art ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a GPU multi-task scheduling system and a method based on Kernel merging. Including task analysis module, Kernel merge module, task scheduling module and task storage queue. Theuser submits the GPU task to the task storage queue, the task analysis module calculates the merging acceleration ratio between tasks, the task scheduling module calculates the optimal merging scheduling sequence, and delivers it to the Kernel merging module for merging execution, so as to obtain the maximum task throughput; The invention adopts the maximum matching of the general graph to calculate the optimal consolidation scheduling order. The task merge acceleration ratio is calculated by task analysis module. Secondly, the directed graph is constructed according to the acceleration ratioof task merging. The optimal task consolidation scheduling order is calculated by bipartite graph matching algorithm. Task groups are sequentially scheduled to complete task calculations. The invention completes the design details of the function, realizes the algorithm and the coding work, and improves the resource utilization rate of the GPU under the multi-task.

Description

technical field [0001] The invention belongs to the field of computer applications, and in particular relates to a GPU task scheduling system and method based on Kernel merging. Background technique [0002] The heterogeneous parallel programming model emerged with the development of the GPU. Since the GPU retained many characteristics of the stream processor, the early heterogeneous parallel programming model inherited the idea of ​​stream programming. Stream is the core of this type of programming model. At the core, a stream is a collection of data, and calculations are executed in parallel on each data element of the stream, which conforms to the execution mode of single instruction multiple data (SIMD) or multiple instruction multiple data (MIMD). In this context, along with the great commercial success of NVIDIA GPU, CUDA appeared in 2007. However, CUDA can only be used in heterogeneous systems with NVIDIA GPUs as acceleration devices. Therefore, at the end of 2008, a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F9/48G06F9/50
CPCG06F9/4881G06F9/5038
Inventor 朱正东田靖轩郭辉李少辉王鹏博韩靖雯李小轩张小雨
Owner XI AN JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com